On the Potential of a Smart Control Valve System for Irrigation Water Network Management ^†

Abstract

Guaranteeing the sustainability of irrigation under the increasing pressures related to resource scarcity and climate change is a major global challenge, specifically in water-scarce areas. The present work proposes a methodological approach and a technological solution to improve operation and management of a critical on-demand pressurized irrigation system. The potential of using an innovative off-grid valve, the green valve, for the real-time monitoring and adaptive control is proven through numerical simulations. Particular emphasis is given to the energy balance feasibility of the system during a typical irrigation season. Reference is made to a real network located in southern Italy.

1. Introduction

Agriculture is a key activity for many countries worldwide, both for economic reasons and for its direct contribution to food security [1]. However, agriculture currently requires around 70% of the world’s freshwater resources and relies heavily on irrigation practices [2]. Additional challenges are related to the effects of climate change and increasing population, which will likely provoke an increase in water used for agriculture, even considering an increase in water use productivity [3].

In this framework, the increase in water use efficiency is crucial to guarantee the sustainability of agriculture over the long term. An increased water use efficiency is tightly connected to the innovation and modernization processes in agriculture, such as, for example, precision agricultural technologies [4] and smart irrigation, which involves the application of the right amount of water at the right time and in the right place [5]. The use of innovative monitoring and control strategies is, therefore, needed for supporting an improved water use efficiency in agriculture, and some reviews already focused on the use of smart technologies such as smart sensors networks, Internet of Things (IoT), and wireless sensor networks (WSN) in agriculture [5]. Much evidence in the literature (e.g., [6,7]) suggests that the use of smart sensors networks can help in different scenarios to control and manage irrigation systems, providing monitoring capabilities and decision-making support, ultimately achieving an improvement in the performance of irrigation practice. Significant savings in water (and energy) consumption can be achieved, along with an easier management of agricultural practices as pumps, motors, and valves, and other devices can be remotely operated.

The mentioned literature also highlights the main features of smart systems used in agriculture. Among them, straightforward communication with the users (including the farmers) should be guaranteed. Furthermore, the response needs to be provided in real-time and rely on a distributed network of sensors/devices [6]. It should be also carefully considered that irrigation districts are complex systems and farmers are agents who have autonomy, intelligence, and relative knowledge about their environment [8].

Within this framework, the present study builds on recent works by the same research group (e.g., [9]), which discusses the potential associated with the introduction of a smart control valve, the green valve (GV) [10,11], in a pressurized irrigation distribution system (PIDS). The GV recovers the energy needed for operation and can create a stand-alone real-time control and monitoring system. It is remotely controlled and can implement management logics for network performance optimization based on the real-time data acquired. The present work deals with a case study (a PIDS located in southern Italy) characterized by the occurrence of pressure deficit in some specific conditions of use. The installation of the GV in several specific nodes of the PIDS is proposed as a solution to reduce the occurrence of pressure deficit. Firstly, the test case is introduced, after which indicators are proposed to evaluate the efficiency of the management solutions implemented with the use of the GV. In the end, the energetic self-sustainability of the GV is discussed.

2. Test Case Description

The present paper considers Sector 25 of District 4 of the “Sinistra Ofanto” irrigation network in Capitanata (Puglia, southern Italy). A scheme of the network is shown in Figure 1. Detailed information on the system and its functioning can be found in some works [12,13]. The pressure at the inlet of the network, i.e., at node #0, changes during the irrigation season depending on the water demand of the other sectors of the district. In particular, it ranges between 128 and 142 m as indicated in [12].

The inlet head and the water demand used in the simulation for each month are reported in

Table 1. Hydrants operation during the irrigation season, water demand, and simulated inlet head conditions.

Contemporary Hydrants	Apr (h)	May (h)	Jun (h)	Jul (h)	Aug (h)	Sept (h)	Oct (h)	Nov (h)
3	15							15
4		10	6			9	12.5
5		10	6	9	10.8	9	12.5
6			8	9	10.8	12
7				12	14.4
Water demand (m3)	540	720	900	1080	1296	1080	900	540
Inlet head (m)	140	136	132	131	128	131	133	139

. Table 1 reports also the working hours of each hydrant in a specific configuration (i.e., set of simultaneous hydrants). For example, in the month of June, a hydrant works for 6 h in configuration with four contemporary hydrants; for 6 h in a configuration with five contemporary hydrants; for 8 h in a configuration with six contemporary hydrants. The number of simultaneous hydrants increases with the increase in water demand, with a maximum of seven contemporary hydrants in the most demanding periods. The assumptions given to reproduce the irrigation season do not aim to exactly reproduce a physical phenomenon or to mimic an observed trend, rather aim to realistically reproduce the randomness of an on-demand PIDS operation, also taking into account some expert knowledge on the operation of this specific sector under investigation.

3. Material and Methods

3.1. Description of the System

In [9], the same test case was analyzed to prove the potentiality of the application of the GV in the most demanding irrigation situation when 7 hydrants are simultaneously open. In particular, the effectiveness of the application of a simple operational rule on the simultaneous use of some critical hydrants, specifically #14 15 16 and #18 19 20 21 22 23 24, was proven. A hydrant is considered critical if the pressure head available at the node is below 20 m. The management rule is implemented installing the GV in the critical nodes and preventing the simultaneous opening of a certain number of hydrants. The present work extends the study, analyzing an entire irrigation season with different demand conditions and inlet pressure heads during the period defined in Section 2.

During an irrigation season, the usage frequency of hydrants changes widely. The application of a single control rule based on the most critical condition is effective in reducing the number of possible critical combinations, as proven in [9], but can be excessively restrictive when the demand is not the highest. To provide a simple example, a single rule can be defined on the basis of the most demanding months (i.e., July and August), but the same rule applied in the other months can be excessively restrictive and prevent the use of hydrants that would not create a critical condition. During months that are not the most demanding, the pressure head at the network inlet is higher allowing for the use of a larger number of hydrants simultaneously, with a reduced probability of pressure deficit occurrence. To reduce the impact of management rule application on users’ access to the resource, the present paper proposes to adapt the management rule on the basis of the pressure head at the inlet node of the network.

The adaptive rules were implemented to obtain at least a 50% reduction in the number of failures with the minimum impact on the availability of hydrants. To take account of these characteristics, the failure reduction rate, $Fr{r}_H$, is defined as:

$$Fr{r}_H=\frac{{{\displaystyle \sum }}_j^{m}f{c}_{a,j}}{{{\displaystyle \sum }}_j^{m}f{c}_{b,j}}·100$$

(1)

where $f{c}_{a,j}$ and $f{c}_{b,j}$ are the number of total failure combination after and before, respectively, the application of the rule in the month $j$, and $m$ is the total number of irrigation months. A minimum failure reduction rate of 50% is required for accepting the rule. A second index was considered to evaluate the impact of rule application on hydrants availability. It is the average stiffness $S_H$ that expresses the average ratio between the number of combinations that the application of the rule inhibits and the total number of possible combinations of active hydrants. The rule defined should minimize this parameter. It is calculated as:

$$S_H=\frac{1}{m} \frac{{{\displaystyle \sum }}_j^m{c}_{i,j}}{{{\displaystyle \sum }}_j^m{c}_j}·100$$

(2)

where $c_{i,j}$ is the number of inhibited combinations and $c_j$ is the total number of possible combinations of active hydrants at month $j$.

3.2. GVS Sizing and Energy Balance

The sizing procedure discussed in [9] was applied for effectively sizing the device. The energy that the GV uses for its operation is calculated considering the effective performance of the GV in the simulated hydraulic conditions during the entire irrigation season. The efficiency of the energy recovery process, from hydraulic to mechanical energy, is calculated on the basis of the effective characteristics of the valve defined in [9]. Instead, a constant efficiency equal to 70% is applied for the conversion from mechanical to electrical energy.

The monthly energy consumption necessary for the sustainability of the system is indicated in

Table 2. Valve operation consumption and monthly demand and usage.

Single Valve	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Avg.
Demand (m3)	540	720	900	1080	1296	1080	900	540	882
Hours of use	15	20	25	30	36	30	25	15	25
Consumption (Wh)	788	833	832	877	902	853	854	788	841

. It was calculated considering the following assumptions about the system: a consumption of 5 W for data communication when the hydrant is open; a consumption of 40 W for 60 s each time the valve is opened and closed due to the actuator supply; an additional consumption due to the valve operation when the hydrant is open. The additional consumption is due to the actuator activation for 3 s each 10 min of use for pressure regulation; finally, a standby consumption of 1 W for the period when the valve is not operating.

The critical irrigation period corresponds to the most demanding months for the device energy consumption because the number of operations is higher and, consequently, the required energy.

4. Results

4.1. Irrigation Season and Operational Rules

A first overall result is that when the inlet pressure head is above 134.8 m, all the possible conditions are not critical, i.e., they do not create pressure deficit conditions. This also applies to the most demanding condition where seven hydrants are open simultaneously. A base rule is defined, as suggested in [9], on the worst condition simulated, i.e., the one in the month of August when the inlet head achieves the minimum value of 128 m.

Table 3. Rule defined on the basis of the system inlet pressure. Each rule achieves a reduction of at least 50% of the failure conditions.

Inlet Head (m)	Rule ID	Max Contemporary in Range: 14–15–16	Max Contemporary in Range: 18–19–20–21–22–23–24
$H_{in}<130$	Base	2	3
$130\le H_{in}<131$	1	3	3
$131\le H_{in}<133$	2	3	3 (#18 excluded)
$133\le H_{in}<134$	3	3	4 (#18 excluded)
$134\le H_{in}<134.8$	4	3	5 (#18 excluded)
$134.8\le H_{in}$	No rule	-	-

provides a summary of the rules applied to the two most critical groups of hydrants. The first one is composed of hydrants #14, #15, and #16, and the second one is composed of hydrants #18, #19, #20, #21, #22, #23, and #24.

Figure 2 shows the two parameters $Fr{r}_H$ and $S_H$ for each value of the head considered in the simulations. Figure 2a shows the average stiffness and Figure 2b shows the failure reduction rate. It can be seen that the $Fr{r}_H$ calculated is always above 50% as required to consider the rule acceptable. Figure 2a shows a significant reduction in the average stiffness, meaning that the effectively realizable combinations of active hydrants is, in the worst case, 26.5% for low values of pressure heads at the inlet of the system, and decrease as a function of the applied rule when the inlet pressure head decreases. As already seen, above 134.8 m not critical condition occurs in the system, making the introduction of a rule unnecessary.

The resulting hydrant arrangement defined through the use of the adaptive rules was used to investigate the energetic sustainability of the GVs during the irrigation season.

4.2. GV Sizing and Energy Balance

In this section, the installation of the GV in the critical nodes of the plant is investigated, with particular attention to the energetic self-sustainability of the system. The sizing procedure detailed in [9] was applied. The energy recovered by the devices is calculated considering the functioning pattern for each month as defined in Table 2. The results are shown in Figure 3, where the boxplot of the monthly recoverable energy by each GV as a function of the month and as a function of the single installation node are reported. The dotted line in Figure 3a indicates the minimum energy necessary for the self-sustainment of the system calculated on a monthly basis as defined in Section 3.2. It can be seen that June and August are critical months, as the values of recovered energy are partially lower than the minimum energy needed for the self-sustainment of the valve during the month. Nevertheless, the average recovered energy for each node showed in Figure 3b is always above the limit of 902 Wh that is the maximum monthly consumption during the season shown in Table 2. This condition indicates a positive balance between energy recovered and consumed during the entire irrigation season for each node where a valve is installed. To prevent a potential lack of power under the most critical operating conditions, and between two subsequent hydrant activations, a battery must be sized in order to accumulate enough energy when the recoverable energy is higher than the consumption.

Moreover, looking at Figure 3b, the most critical nodes are those showing the lowest values of recovered energy, i.e., nodes #18, #19, #20, #22, and #23. These nodes must be equipped with a larger battery to cover energy deficit periods during the months of July and August.

5. Discussion and Conclusions

The present work builds on an active research line of the group of authors, which is basically oriented to support an increased efficiency and improved operation and management of on-demand PIDS. The approach proposed is based on the use of a set of GVs, working off-grid, for the real-time monitoring of operating conditions and remote adaptive PIDS control based on a set of simple rules that should reduce hydrant failures due to pressure deficit. Reference is made to a PIDS located in southern Italy. The main added value of this work lies in the feasibility assessment of the GV system installation, based on the simulation of its behavior throughout the irrigation season. Particular emphasis is given to the GV energy balance, which can be particularly critical for irrigation applications.

First, it should be highlighted that the definition of simple adaptive rules during the irrigation season allows for a significant reduction in the number of critical hydrant combinations, also reducing the impact on hydrant availability with respect to a single rule application. The proper design and installation of a network of GVs on a PIDS can, therefore, allow a safe operation of the irrigation system, while ensuring the continuity of water supply and the on-demand operation of the system.

Secondly, it should be noted that the potential of GV introduction is directly related to the capability of having real-time monitoring and management of the PIDS. In line with the principles of ‘smart’ irrigation, the GV system may provide a more rational and efficient water use, which can positively impact both the water managers and the users.

Lastly, it should be remarked that, besides being self-sustainable from an energetic point of view, the GV may also allow an additional recovery/production of energy (depending on the period of the year and on the actual operating conditions). This aspect has manifold potential implications, starting, e.g., from the capacity to support an additional monitoring of water quality parameters.

Although the advantages and key implications of the introduction of a system of GVs on a PIDS are shown, some issues need specific attention before a wide uptake of this technological solution. An economic analysis is a direct development of the current study. The evidence of the technical advantages that the use of this kind of devices can contribute to the operation in pressurized on-demand irrigation systems is apparent, and an economic analysis would provide better understanding of the financial implications and the medium-term sustainability of its implementation in real systems. In this direction, understanding the ‘willingness to pay’ of the key stakeholder would be crucial to build a reliable business model for the systems. This would be also directly related to the ‘bottom-up’ identification of the key additional technical features (e.g., specific parameters to be monitored) that would make the proposed technological solution marketable. Field tests are also foreseen in the future, with the aim of both testing the real operation of the GV in irrigation applications, and of understanding and valuing its key features for the final users.