An artificial bee colony algorithm-based water resource optimal allocation method for irrigation areas

Abstract

The present application relates to the field of wisdom agricultural water conservancy technology, and discloses a method for optimizing allocation of water resources in irrigation areas based on artificial bee colony algorithm, comprising the following steps: step 1: obtaining basic information of water resources in the irrigation area; step 2: establishing an optimized water distribution model under multiple constraint conditions with the maximum economic benefit of the irrigation area as the target; and step 3: solving the model, first initializing the number of populations and the number of honey sources, the quality of the honey source being equivalent to the solution of the function, evaluating the fitness of the individual with the quality of the solution, determining the type of bees according to the quality of the solution, leading the bees to search for new honey sources, replacing the old ones if better, and abandoning the old ones if a honey source within the threshold value cannot find a better one. The scout bees find better honey sources. When the number of iterations reaches the preset value, the iteration is stopped. The present application can quickly obtain the global optimal solution of the benefit of the irrigation area, reasonably allocate the irrigation amount of groundwater and surface water, promote the sustainable development of ecology, and optimize and adjust the planting area distribution of various crops in the irrigation area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent agricultural water conservancy technology, in particular to a method for optimizing allocation of water resources in irrigation areas based on artificial bee colony algorithm. BACKGROUND

[0002] The development of irrigation areas cannot be separated from the efficient use of water resources. China is a country with a shortage of water resources per capita. The total amount of water resources is abundant, about 2800 billion cubic meters, but the available rate is less than half, and the distribution is uneven. Due to the large population base in China, the amount of water resources per capita is very low. In recent years, the total amount of water supply in China is about 550 billion cubic meters, and the amount of irrigation water accounts for more than half of the total water supply, about 360 billion cubic meters, but the loss rate of agricultural irrigation reaches 60%. With the improvement of people's living standards and the development of industrial society, the total amount of domestic and industrial water is also increasing, which will inevitably reduce the proportion of agricultural water. In order to maximize the benefits of irrigation areas and make the ratio of groundwater and surface water more reasonable, promote the sustainable development of water resources, it is crucial to comprehensively use computer technology, artificial bee colony algorithm technology to optimize the allocation of water resources in irrigation areas.

[0003] Looking back at previous studies, the optimization of water resources in irrigation areas is mostly concentrated on the optimization of water distribution in the canal system, single target optimization of water distribution, and the commonly used algorithms are traditional genetic algorithm, particle swarm algorithm, etc. Among them, the advantages of genetic algorithm are strong search ability and scalability, but it is strongly dependent on the quality of the initial population, and the parameters contained in the algorithm need to be determined by experience, which greatly reduces the accuracy of the optimized water distribution. The particle swarm algorithm has the advantages of simple concept, easy implementation and strong universality, but the robustness of the algorithm is poor, and the search performance is dependent on the parameters to a certain extent, and it cannot track the optimal solution well in a dynamic environment. Artificial bee colony algorithm has no special requirements for optimization function and initial value, and has strong global search ability and strong universality, and is suitable for solving dynamic complex models for optimizing water distribution in irrigation areas based on water demand. Therefore, the present application provides a method for optimizing allocation of water resources in irrigation areas based on artificial bee colony algorithm. SUMMARY

[0004] The present application provides a method for optimizing allocation of water resources in irrigation areas based on artificial bee colony algorithm, which can solve the problem of unreasonable water distribution and groundwater consumption in irrigation areas and the lack of maximization of irrigation area benefits in the prior art.

[0005] To solve the above problems, the present application is realized by the following technical scheme:

[0006] Step 1: First, the basic information of water resources in the irrigation area is obtained, including the total amount of surface water resources in the irrigation area, the surface water taking cost in the irrigation area, the total amount of groundwater resources in the irrigation area, and the groundwater exploitation cost in the irrigation area; the basic information of crops in the irrigation area, including the potential evapotranspiration of each crop, the irrigation area, the sales unit price and the unit area yield, the crop water requirement, and the irrigation water requirement.

[0007] Step 2: Considering the crop evapotranspiration and effective rainfall, an optimization water allocation model under multiple constraints is established with the goal of maximizing economic benefits in the irrigation area.

[0008] Step 3: The model is solved, first initializing the number of populations and the number of honey sources, the quality of the honey source is equivalent to the solution of the function, the fitness of the individual is evaluated by the quality of the solution, and the quality of the solution is determined according to the quality of the solution, the first 50% of the solution is the leading bee, and the last 50% of the solution is the following bee.

[0009] Step 4: Update of new honey source, when the leading bee finds a new honey source with better fitness than the original honey source, replace the original honey source with the new one, enter the next generation, and form a new leading bee population and honey source position.

[0010] Step 5: Following the search, calculate the probability of the honey source i being selected, i.e. the probability of the following bee selecting the leading bee, if the selected honey source is better, replace the old honey source.

[0011] Step 6: Produce scout bees, during the search process, if a honey source has not been updated to a better one after n iterations of search reaching the threshold L, a scout bee will be produced, which will search for a better honey source.

[0012] Step 7: Determine whether the iteration number is less than the preset maximum value, if yes, go to step 4; otherwise, output the optimal solution.

[0013] Step 1 above, effective rainfall (PE) is the rainfall stored in the root zone, which can be easily obtained and utilized by crops, generally calculated using the empirical rainfall utilization coefficient, the formula is as follows:

[0014] PE = αP

[0015] Where P is the rainfall, and α is the rainfall utilization coefficient.

[0016] The potential evapotranspiration of crops is usually calculated using empirical equations based on weather data, using the Hargreaveamani formula to calculate the potential evapotranspiration of crops, the formula is as follows:

[0017]

[0018] Where PET is the potential evapotranspiration (mm / day), T meanis the daily average temperature (℃), T max is the daily maximum temperature (℃), T min is the daily minimum temperature (℃), R a is the total solar radiation (mm / day).

[0019] Crop water requirement is the amount of water required by a crop to meet its normal growth during a given growth period through evapotranspiration loss, and its calculation formula is as follows:

[0020] ET c = K c * PET

[0021] wherein, ET c is the crop water requirement, K c is the crop transpiration coefficient, and PET is the crop potential transpiration.

[0022] Irrigation water requirement (IWR) is the total amount of water supplied through irrigation infrastructure to meet the water requirement during the crop water shortage period, and the irrigation water requirement is calculated from the crop water requirement and effective rainfall, and the formula is as follows:

[0023] IWR = ET c - PE

[0024] wherein, IWR is the irrigation water requirement, ET c is the crop water requirement, and PE is the effective rainfall.

[0025] In the above step 2, the crop evapotranspiration and effective rainfall are considered, and the objective function is:

[0026]

[0027] The constraint conditions include:

[0028] (1) Surface water supply constraint:

[0029] (2) Water balance constraint:

[0030] (3) Total water resource constraint:

[0031] (4) Crop planting area constraint:

[0032] wherein, F is the economic benefit of the irrigation area, j indicates the crop type, C j indicates the price of j crops, A j is the area of j crops, Y j is the unit area yield of j crops, and wS jThe amount of surface water supplied to the crop j, cW is the cost of surface water, cG is the cost of groundwater, wS is the surface water supply, gS is the groundwater supply, IWR j The irrigation water requirement of j crops, ET ij The potential evapotranspiration of j crops in the ith month, PE i The effective rainfall in the ith month, A is the total area of the irrigation area.

[0033] In the above step 3, initialize the bee colony, set the bee colony size N, the number of nectar sources M, the maximum number of iterations T, and the threshold value L. Randomly allocate groundwater or surface water, but prefer to use surface water, which is abundant and cheap, and saving the use of groundwater helps promote sustainable development of water resources, and the water resource ratio is 0.033, the calculation formula of the nectar source position is as follows:

[0034] A(i, j) = A min (j) + rand(1)*(A max (j)-A min (j))

[0035] wS(i, j) = rand(1)*S(j); gS(i, j) = 0.033*rand(1)*S(j)

[0036] Where A i (i = 1, 2…, X), wS i (i = 1, 2…, X), gS i (i = 1, 2…, X) is the ith nectar source position in the population, X is the number of nectar sources, A(j) min and A(j) max is the boundary of A i (i = 1, 2…, X) in the jth direction, rand(1) is a uniformly distributed random number in [0, 1].

[0037] The calculation formula of the fitness of the nectar source is as follows:

[0038]

[0039] Where f(x i ) is the objective function value of the problem to be solved:

[0040]

[0041] In the above step 4, the position of the new nectar source is determined as follows:

[0042]

[0043] Where is a uniformly distributed random number, which determines the degree of disturbance, a* is an acceleration coefficient, usually taking 1.

[0044] In the above step 5, the probability of the honey source i being selected is p i :

[0045]

[0046] wherein the value of the fitness of the i th solution is fit i , p i is the proportion of the yield of the honey source in the whole bee colony, and the larger p i is, the better the quality of the honey source is, and the higher the quality of the feasible solution is.

[0047] In the above step 6, the position of the new honey source found by the scout bee is:

[0048] A(i,j)=A min (j)+rand(1)*(A max (j)-A min (j))

[0049] wS(i,j)=rand(1)*S(j);gS(i,j)=0.033*rand(1)*S(j)

[0050] If the new search new honey source is better than the old honey source position, according to the greedy rule, the new honey source will replace the old honey source.

[0051] Compared with the prior art, the beneficial effects of the present application are that the irrigation district water resource optimal allocation method based on the artificial bee colony algorithm effectively improves the economic benefits of the irrigation district, meanwhile makes the irrigation district water resource allocation more scientific, reduces the exploitation amount of groundwater, and optimizes the crop planting structure of the irrigation district. The artificial bee colony algorithm adopted has the advantages of strong robustness, few control parameters, simple calculation, and easy implementation, and can quickly obtain the optimal solution of the model. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a flow chart of the irrigation district water resource optimal allocation method based on the artificial bee colony algorithm provided by the embodiment of the present application.

[0053] Figure 2 is a schematic comparison diagram of the running iteration process of the artificial bee colony algorithm and the particle swarm algorithm for solving the optimal value in the embodiment of the present application. Figure 1 DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the present application more clear and clear, the present application is further described in combination with specific examples.

[0055] ​An artificial bee colony algorithm-based irrigation water resource optimal allocation method, as shown in the formula (I) comprises the following steps: Figure 1

[0056] Step 1: First, obtain the basic information of the irrigation water resources, including the multi-year continuous effective rainfall of the irrigation area, the total amount of surface water resources in the irrigation area, the surface water taking cost of the irrigation area, the total amount of groundwater resources in the irrigation area and the groundwater exploitation cost of the irrigation area; the basic information of crops in the irrigation area, including the crop potential evapotranspiration of each crop, the irrigation area, the sales unit price and the unit area yield, the crop water requirement and the irrigation water requirement.

[0057] The crop effective rainfall (PE) is the rainfall stored in the root zone, which can be easily obtained and utilized by crops, and is generally calculated by using an empirical rainfall effective utilization coefficient, and the formula is as follows:

[0058] PE = αP

[0059] Wherein, P is the rainfall, and α is the rainfall effective utilization coefficient.

[0060] The crop potential evapotranspiration is usually calculated according to weather data by using an empirical equation, and the Hargreaves equation is used to calculate the crop potential evapotranspiration, and the formula is as follows:

[0061]

[0062] Wherein, PET is the potential evapotranspiration (mm / day), T mean is the daily average temperature (℃), T max is the daily maximum temperature (℃), T min is the daily minimum temperature (℃), and R a is the total solar radiation (mm / day).

[0063] The crop water requirement is the water loss through evapotranspiration in a given growth period to meet the water requirement of normal growth of crops under given environmental conditions, and the calculation formula is as follows:

[0064] ET c = K c * PET

[0065] Wherein, ET c is the crop water requirement, K c is the crop transpiration coefficient, and PET is the crop potential evapotranspiration.

[0066] The irrigation water requirement (IWR) is the total water supplied through irrigation infrastructure to meet the water requirement during the water shortage period of crops, and the irrigation water requirement is calculated from the crop water requirement and the effective rainfall, and the formula is as follows:

[0067] IWR = ET c ​- PE

[0068] Where, IWR is irrigation water requirement, ET c is crop water requirement, PE is effective rainfall.

[0069] Step 2: Considering crop evapotranspiration and effective rainfall, an optimal water distribution model under multiple constraints is established to maximize the economic benefit of the irrigation district.

[0070] The objective function is:

[0071]

[0072] The constraints include:

[0073] (1) Surface water supply constraint:

[0074] (2) Water balance constraint:

[0075] (3) Total water resources constraint:

[0076] (4) Crop planting area constraint:

[0077] Where, F is the economic benefit of the irrigation district, j refers to the crop type, C j refers to the price of j crops, A j is the area of j crops, Y j is the unit area yield of j crops, wS j is the amount of surface water supplied to crop j, cW is the cost of surface water, cG is the cost of groundwater, wS is the surface water supply, gS is the groundwater supply, IWR j is the irrigation water requirement of j crops, ET ij is the potential transpiration of j crops in the ith month, PE i is the effective rainfall in the ith month, A is the total area of the irrigation district.

[0078] Step 3: The artificial bee colony algorithm is used to solve the model. First, initialize the number of populations and the number of honey sources. The quality of the honey source is equivalent to the solution of the function. The quality of the solution is used to evaluate the fitness of the individual. The quality of the solution determines the type of bee. The first fifty percent of the solution is the leading bee, and the last fifty percent of the solution is the following bee.

[0079] Initialize the bee colony, set the bee colony size N to 200, the number of honey sources M to 100, the maximum number of iterations T to 500, and the search problem dimension D to 4. The calculation formula of the honey source position is as follows:

[0080] x ij = xmin j +rand[0,1](x max j -x min j )

[0081] where x i (i = 1, 2…, X) is the i-th honey source location in the population, X is the number of honey sources, x min j and x max j are the boundaries of x i (i = 1, 2…, X) in the j-th direction, rand[0,1] is a random number uniformly distributed in [0,1]. In the specific implementation process, four kinds of area combinations are randomly generated, and the amount of groundwater and surface water is randomly allocated:

[0082] A(i, j) = A min (j) + rand(1)*(A max (j) - A min (j))

[0083] wS(i, j) = rand(1)*S(j); gS(i, j) = 0.033*rand(1)*S(j)

[0084] In order to better balance groundwater and surface water, promote the sustainable development of water resources, preferentially use surface water, and the proportion of water resources is 0.033.

[0085] The fitness calculation formula of the honey source is as follows:

[0086]

[0087] In the specific implementation, the fitness of the initial honey source is the objective function:

[0088]

[0089] Step 4: update of new honey source, when the fitness of the new honey source is better than the original honey source, replace the original honey source with the new honey source, enter the next generation, and form a new leading bee population and honey source location.

[0090] The position determination formula of the new honey source is as follows:

[0091]

[0092] where is a uniformly distributed random number, which determines the degree of disturbance, a * is the acceleration coefficient, usually taking 1.

[0093] The formula for the foraging bee to search for crop area is as follows:

[0094] VA(i, j) = A(i, j)(2*rand(l)-l)(A(i, j)-A(k, j)) min

[0095] The formula for the distribution of surface water and groundwater for the employed bees is as follows:

[0096] VwS(i, j) = wS(i, j) + (2*rand(l)-l)(wS(i, j)-wS(k, j))

[0097] VgS(i, j) = gS(i, j) + (2*rand(l)-l)(gS(i, j)-gS(k, j))

[0098] Step 5: Following the bees search, the probability of the selected honey source i, i.e. the probability of the following bees selecting the scout bees, is calculated. If the selected honey source is better, it replaces the old honey source.

[0099] The probability of the selected honey source i is p i :

[0100]

[0101] where the fitness value of the ith solution is fit i , and p i is the proportion of the yield of the honey source in the entire bee colony. The larger p i is, the better the quality of the honey source and the feasible solution is. Further, the calculation of p i uses the roulette wheel algorithm, also known as the proportional selection method, whose basic idea is that the probability of selection of an individual is proportional to its fitness.

[0102] Step 6: Generate scout bees. During the search process, if a honey source has not been updated to a better one after n iterations of search reaching the threshold value L, a scout bee will be generated. The scout bee will search for a better honey source.

[0103] The position of the new honey source found by the scout bee is:

[0104]

[0105] If the newly searched new honey source is better than the old honey source position, according to the greedy rule, the new honey source will replace the old one. In specific implementation, the formula for the scout bee to find the optimal area and the optimal distribution of surface water and groundwater is as follows:

[0106] A(i, j) = A min (j) + rand(l)(A max (j)-A min (j))

[0107] wS(i,j) = rand(l) * S(j); gS(i,j) = 0.033 * rand(l) * S(j)

[0108] Step 7: judging whether the iteration number is less than the preset maximum value, if yes, turning to step 4; otherwise, outputting the optimal solution.

[0109] Embodiment

[0110] The selected embodiment of the present application is the irrigation area data of Zaohe Irrigation Area in Suqian City, Jiangsu Province.

[0111] Zaohe Irrigation Area is located in the northwest of Suqian City, Jiangsu Province, and borders on Luoma Lake and Beijing-Hangzhou Grand Canal, providing agricultural irrigation and drainage services for Suqiyu District, Suqian District and Suqian Development Zone. The total control area of the irrigation area is 3.39 million mu, and the designed irrigation area is 2.29 mu. This place belongs to the warm temperate monsoon climate, and the annual precipitation is around 1000 mm, which is unevenly distributed, mostly concentrated in June to August.

[0112] Table 1 is the basic information of crops in Zaohe Irrigation Area:

[0113] Table 1 Basic information of crops

[0114]

[0115] Table 2 is the annual distribution of multi-year average precipitation in the irrigation area:

[0116] Table 2 Annual distribution of multi-year average precipitation

[0117]

[0118] Table 3 is the evapotranspiration of crops in different growth periods:

[0119] Table 3 Evapotranspiration of different crops

[0120]

[0121] The available amount of water resources in the irrigation area is about 240 million cubic meters, the annual exploitation amount of groundwater is 7 million cubic meters, and the surface water supply is 233 million cubic meters. The price of surface water is 0.3 yuan per cubic meter, and the price of groundwater is 0.8 yuan per cubic meter.

[0122] After using the artificial bee colony algorithm for solution, the optimal water allocation results of each crop are shown in Table 4:

[0123] Table 4 Optimal water allocation results of crops

[0124]

[0125] The planting area of various crops after optimization is shown in Table 5:

[0126] Table 5 crop area optimization results

[0127]

[0128] At this time the model corresponding to the objective function value is 8.8*10 8 Economic benefits of the irrigation area are improved, and crop evapotranspiration and effective precipitation are considered, so that water resources and crop planting structure of the irrigation area can be optimized simultaneously. Figure 2 As shown in the iteration curve comparison between the artificial bee colony algorithm and the particle swarm algorithm, the results obtained by running the program ten times are the same, and the running time is less than 2 seconds.

[0129] The above disclosure is only one specific embodiment of the present application, but the embodiments of the present application are not limited thereto, and any changes that can be thought of by those skilled in the art shall fall within the protection scope of the present application.

Claims

1. An artificial bee colony algorithm-based water resource optimal allocation method for irrigation areas, characterized in that, The method comprises the following steps: Step 1: first, the basic information of the crops and irrigation water resources in the irrigation area is acquired; Step 2: an optimal water distribution model under multiple constraint conditions is established with the maximum economic benefit of the irrigation area as the target, and the crop evapotranspiration and effective rainfall are considered, and the target function is: The constraint conditions include: (1) Available surface water volume constraint: (2) Water balance constraint: (3) Total water resource constraint: (4) Crop planting area constraints: Where F is the economic benefit of the irrigation area, j refers to the crop type, C j refers to the price of the jth crop, A j is the area of the jth crop, Y j is the yield per unit area of the jth crop, wS j is the amount of surface water supplied to the jth crop, cW is the cost of surface water, cG is the cost of groundwater, IWR j is the irrigation water requirement of the jth crop, wS is the surface water supply, gS is the groundwater supply, ET ij is the potential evapotranspiration of the jth crop in the ith month, PE i is the effective rainfall in the ith month, A is the total area of the irrigation area, A jmax is the maximum area that the jth crop can be planted, A jmin is the minimum area that the jth crop can be planted, η is the utilization rate of irrigation water resources; Step 3: the model is solved, first, the number of populations N and the number of honey sources M, the maximum number of iterations T, the threshold value L are initialized, the quality of the honey source is equivalent to the solution of the function, the fitness of the individual is evaluated by the quality of the solution, the bee species is determined according to the advantages and disadvantages of the solution, the first fifty percent of the solution is the leading bee, and the last fifty percent of the solution is the following bee; Step 4: the update of the new honey source, when the fitness of the new honey source is better than that of the original honey source, the new honey source replaces the original honey source, enters the next generation, and constitutes a new leading bee population and honey source position; Step 5: the following bee searches, the probability that the honey source i is selected is calculated, that is, the probability that the following bee selects the leading bee, if the selected honey source is better, the old honey source is replaced; Step 6: the scout bee is generated, in the search process, if a honey source reaches the threshold value L after n times of iteration search and has not been updated to a better honey source, the scout bee will be generated, and the scout bee will search for a better honey source; Step 7: it is judged whether the number of iterations is less than the preset maximum value, if yes, go to step 4; otherwise, the optimal solution is output.

2. The method according to claim 1, wherein, In the step 1, the basic information of the crops and irrigation water resources in the irrigation area includes the effective rainfall in the irrigation area for many years in succession, the total amount of surface water resources in the irrigation area, the surface water taking cost in the irrigation area, the total amount of groundwater resources in the irrigation area and the groundwater exploitation cost in the irrigation area; the basic information of the crops in the irrigation area includes the crop potential evapotranspiration of each crop, the irrigation area, the sales unit price and the unit area yield, the crop water requirement and the irrigation water requirement.

Images