Energy safety collaborative early warning method for agricultural park
By establishing a time-scale decomposition model and constructing a model reliability model, the problem of the separation of the agricultural environmental monitoring system and the energy system is solved, the time-scale division is optimized and the disturbance resistance of the coupled system is evaluated, and the effectiveness of coordinated early warning of energy security is improved.
Patent Information
- Application Number
- CN202510070092.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-16
- Publication Date
- 2025-05-13
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The agricultural environmental monitoring system and the energy system are separated from each other in operation and management, and lack linkage mechanisms and unified management, which leads to the inability to deal with common problems of chain risks such as facility agricultural power outages and agricultural power load overloads.
By establishing a time-scale decomposition model, sensitive error information and decomposition difference coefficients are obtained, coupled with coupled fluctuation information and synergistic load coefficients, model reliability models are constructed, model robustness of energy security synergistic warnings are evaluated, and model robustness index is generated.
The accuracy of time scale division is optimized, the ability of the coupled system to resist disturbed inputs is evaluated, and the effectiveness of coordinated warning and the accuracy of risk analysis is improved.
Smart Images
Figure CN119991025A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of agricultural parks, and more specifically, to a collaborative early warning method for energy security in agricultural parks. Background Art
[0002] The agricultural environmental monitoring system and the energy system are separated from each other in operation and management, lacking linkage mechanism and unified control, making it impossible to cope with common problems of chain risks such as power outages in facility agriculture and agricultural power overload. The existing collaborative early warning of energy security in agricultural parks, by integrating relevant theories of agricultural science, information science and power science, expands big data intelligent technology, and designs an operation safety collaborative early warning method that integrates coupled behavior modeling, risk linkage measurement and collaborative early warning analysis. In the process of collaborative early warning of operation safety, modeling based on multi-time scale decomposition and coupling is the basis of risk linkage measurement and collaborative early warning analysis.
[0003] Modeling based on multi-time scale decomposition and coupling involves different fields of agriculture and energy. The data structures and collection methods of different fields are different, which makes it difficult to directly integrate the data. The operating environment and dynamic change speed of agricultural and energy systems are different, and the mutual influence between fast and slow systems is difficult to capture. At the same time, the interaction relationship between cross-systems is nonlinear and complex, which makes it impossible to effectively express the traditional modeling method. Inefficient modeling leads to unstable sensitivity to risk linkage and early warning analysis, which in turn affects the effectiveness of energy security coordination.
[0004] On the other hand, the capture of nonlinear interactions between cross-domain systems is based on the model of time scale decomposition. The effectiveness and accuracy of time scale decomposition must match the logic of risk linkage, so that the effectiveness of collaborative early warning analysis can be guaranteed. While improving the time scale decomposition capability, evaluating the matching of the effectiveness of risk linkage and the accuracy of time scale division is an urgent problem to be solved.
[0005] In order to solve the above defects, a technical solution is now proposed. Summary of the invention
[0006] The purpose of the present invention is to provide a collaborative early warning method for energy security in an agricultural park to address the deficiencies in the background technology.
[0007] In order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for coordinated early warning of energy security in an agricultural park, the specific steps of which include:
[0008] Obtain sensitive error information based on the time scale decomposition model used to decompose the agricultural system and the energy system, and obtain the decomposition difference coefficient based on the sensitive error information;
[0009] According to the risk linkage relationship of energy security coordination, coupling fluctuation information is obtained, and the coordination load coefficient is obtained according to the coupling fluctuation information;
[0010] The modeling confidence model is constructed based on the decomposition difference coefficient and the collaborative load coefficient. The modeling robustness of the energy security collaborative early warning is evaluated by the ability of the coupling term to affect the fluctuation of the time scale decomposition model, and the model robustness index is generated.
[0011] The model performance is graded according to the time cost and computing cost of the modeling credibility model analysis, a baseline robustness threshold is preset, and the model robustness index is compared with the baseline robustness threshold, and the stability of the modeling credibility model is classified according to the comparison results;
[0012] Based on the grading and classification results, an early warning verification intervention strategy is generated for the effectiveness of the collaborative early warning and notified to the management personnel.
[0013] Preferably, the method for constructing the time scale decomposition model for decomposing the agricultural system and the energy system is:
[0014] A time scale decomposition model for decomposing agricultural systems and energy systems is established. The fast system in the time scale decomposition model is defined to include environmental parameters and consumption parameters. Environmental parameters include temperature T, humidity H, and light intensity I. Consumption parameters include power consumption P. f 、Energy load L f The slow system in the time scale decomposition model includes growth parameters and allocation parameters. Growth parameters include growth stage G(t) and land nutrients N, and allocation parameters are energy reserves E. s ;
[0015] The fast system equation is established as In the formula, X f is the parameter variable of the fast system, and the equation of the slow system is established as In the formula, X s is the parameter variable of the slow system, and the coupling term is set to G(X f ,X s ) represents the linkage relationship between the fast system and the slow system. The fast system after linkage is The slow system after linkage is
[0016] Preferably, the method for obtaining sensitive error information according to the time scale decomposition model is:
[0017] Obtain the sensitivity response coefficients of the time scale decomposition model within the periodic time Q to the input parameters of the fast system and the slow system, and integrate the sensitivity response coefficients of the time scale decomposition model within several periodic time Q to the input parameters of the fast system and the slow system into a short-term set and a long-term set, marking the short-term set as Ssy = {sr p}, where p = {1, 2, 3, ..., o}, and o is a positive integer, sr p It represents the sensitivity response coefficient of the time scale decomposition model within the pth period time Q to the slow system input parameter. The calculation expression of the sensitivity response coefficient is: Where T represents the time scale, and the long-term set is Lsy={lr u}, where u = {1, 2, 3, ..., y}, and y is a positive integer, lr u It represents the sensitivity response coefficient of the time scale decomposition model within the u-th cycle time Q to the slow system input parameters. The calculation expression of the sensitivity response coefficient is: Where G represents the coupling term;
[0018] Calculate the average standard deviation of the sensitivity response coefficient of the time scale decomposition model to the input parameters of the fast system and the slow system within several cycle times Q. The calculation expression is: Where avg1 is the average value of the sensitivity response coefficient of the time scale decomposition model to the input parameters of the fast system within several cycle times Q. The calculation expression is: avg2 is the average value of the sensitivity response coefficient of the time scale decomposition model to the slow system input parameters within several cycle times Q. The calculation expression is:
[0019] Preferably, the method for obtaining the decomposition difference coefficient through sensitive error information is:
[0020] The gradient descent algorithm is used to adjust the weight coefficient of the coupling term, the weight coefficient of the coupling term is marked as θ, and the coupling term is defined as G(X f ,X s ,θ)=θ·f(X f ,X s ), where f is the coupling function between the fast system and the slow system, and the difference between the output value after the coupling and the actual value is calibrated In the formula, n is the number of samples, i is the sample index, is the output value after coupling, Y i is the actual value, L(θ) is the loss function, and the gradient of the loss function to the weight is calculated by derivation. The calculation expression is: In the formula, the difference between the weight update and the weight update is calculated by gradient descent, and the expression is In the formula, α is the learning rate, and the calculation expression of the decomposition difference coefficient is Where D dc is the coefficient of decomposition.
[0021] Preferably, the method for obtaining coupling fluctuation information is:
[0022] Obtain the abnormal response coefficient of the data interaction between the agricultural system and the energy system within the cycle time T, and integrate the abnormal response coefficients of the data interaction between the agricultural system and the energy system within several cycle times T into a data set, and mark the data set as E xr ={in d}, where d = {1, 2, 3, ..., f}, and f is a positive integer, in d It represents the data interaction abnormal response coefficient between the agricultural system and the energy system within the dth cycle time T. The calculation expression of the data interaction abnormal response coefficient is: Among them, t h is the energy warning response time, t j The energy warning will stop reporting time. ae Data transmitted from the agricultural system to the energy system, D ea data transmitted from energy systems to agricultural systems;
[0023] Calculate the standard deviation of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T. The calculation expression is: Where res is the average value of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T, and the calculation expression is:
[0024] Preferably, the method for obtaining the cooperative load factor according to the coupling fluctuation information is:
[0025] The system equation for calibrating the coupling relationship between the agricultural system and the energy system is Z, which is expressed as Among them, X f is the input variable of the agricultural system, X s is the input variable of the energy system, F(X f , X s ) is the nonlinear coupling equation of the agricultural system, S(X f , X s ) is the nonlinear coupling equation of the energy system. The numerical integration method is used to calculate the Lyapunov index of the system equation Z. The calculation expression is: Among them, Z 0 is the initial state of the system equation, δ is a very small constant, indicating a state with a small difference from the initial state, and δ is a positive number, t is the time, then the calculation expression of the collaborative load coefficient is C of =C om λ+1 .
[0026] Preferably, the logic of constructing the modeling reliability model according to the decomposition difference coefficient and the collaborative load coefficient is:
[0027] The expression of the modeling confidence model is: Where M re is the model robustness index, F re is the frequency of data transmission between the energy system and the agricultural system within the cycle time T.
[0028] Preferably, the logic for grading model performance according to the time cost and computing power cost of modeling credibility model analysis is:
[0029] The computation time for calibrating the modeling confidence model is C pt , the energy consumption of the modeling credibility model is E co , the model performance index is calculated as Preset the first performance threshold M f and the second performance threshold M s , and the first performance threshold M f Less than the second performance threshold M s , when the calculated model performance index M dp Less than or equal to the first performance threshold M f When the model performance is defined as M1, the calculated first performance threshold M f Less than the model performance index M dp Less than the second performance threshold M s When the model performance is defined as M2, the calculated model performance index M dp Greater than or equal to the second performance threshold M s When , the model performance is defined as M3 level.
[0030] Preferably, the logic for classifying the stability of the modeling confidence model is:
[0031] Preset benchmark robust threshold B as , the model robustness index M re With the benchmark robust threshold B as When the calculated model robustness index M re Greater than or equal to the benchmark robust threshold B as When , the modeling confidence model is marked as sensitive. When the calculated model robustness index M re Less than the benchmark robustness threshold B as When , the marker modeling reliability model is hysteresis type.
[0032] Preferably, the logic of generating an early warning verification intervention strategy for the effectiveness of the collaborative early warning based on the grading results and the classification results, and notifying the management personnel is:
[0033] When the model performance of the modeling confidence model is at level M1 and the model type is sensitive, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is large, the short-term warning credibility is high, and the long-term warning credibility is low;
[0034] When the model performance of the modeling confidence model is at level M2 or M3, and the model type is sensitive, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is large, the short-term warning credibility is medium, and the long-term warning credibility is low;
[0035] When the model performance of the modeling confidence model is at level M1 and the model type is hysteresis type, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is small, the short-term warning credibility is high, and the long-term warning credibility is high;
[0036] When the model performance of the modeling confidence model is at level M2 or M3, and the model type is hysteresis type, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is small, the short-term warning credibility is low, and the long-term warning credibility is medium.
[0037] In the above technical solution, the technical effects and advantages provided by the present invention are:
[0038] This application decomposes the time scale of the agricultural system and the energy system by establishing a time scale decomposition model, and obtains sensitive error information, evaluates the influence of input variables on output results without being disturbed by the time scale, avoids the situation where the influence of input variables is underestimated or overestimated during the time scale division process, and thus optimizes the overall accuracy of the time scale division. By obtaining the decomposition difference coefficient, the coupling system based on the energy system and the agricultural system is evaluated for its ability to resist disturbance inputs, and the correlation analysis effect of the coupling system is tested. The decomposition difference coefficient and the collaborative load coefficient are comprehensively evaluated through the modeling credibility model, and a verification model for the modeling analysis of the nonlinear coupling of the energy system and the agricultural system is obtained. The ability to effectively refine the linkage influence of nonlinear fluctuations and fluctuations on the energy system and the agricultural system is verified, which provides verification for the effectiveness of the early warning collaborative analysis. By decomposing from the perspective of fast and slow systems and then coupling from the perspective of energy and agriculture, the difference between the decomposition scale and the coupling scale is effectively resolved, and the degree of fusion of the coupling system is further improved, thereby optimizing the accuracy of the correlation risk analysis and the effectiveness of the collaborative early warning. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0040] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION
[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0042] Example 1: Please refer to Figure 1 As shown, the present invention is a collaborative early warning method for energy security in an agricultural park, and the specific steps include:
[0043] Obtain sensitive error information based on the time scale decomposition model used to decompose the agricultural system and the energy system, and obtain the decomposition difference coefficient based on the sensitive error information;
[0044] According to the risk linkage relationship of energy security coordination, coupling fluctuation information is obtained, and the coordination load coefficient is obtained according to the coupling fluctuation information;
[0045] The modeling confidence model is constructed based on the decomposition difference coefficient and the collaborative load coefficient. The modeling robustness of the energy security collaborative early warning is evaluated by the ability of the coupling term to affect the fluctuation of the time scale decomposition model, and the model robustness index is generated.
[0046] The model performance is graded according to the time cost and computing cost of the modeling credibility model analysis, a baseline robustness threshold is preset, and the model robustness index is compared with the baseline robustness threshold, and the stability of the modeling credibility model is classified according to the comparison results;
[0047] Based on the grading and classification results, an early warning verification intervention strategy is generated for the effectiveness of the collaborative early warning and notified to the management personnel.
[0048] Analyze the agriculture-energy spatial mapping relationship, establish a facility agriculture light-heat-water environmental control load model that represents the deep coupling of the agricultural system and the energy system, so as to solve the problem that cross-domain operation safety coupling behavior is difficult to integrate modeling and correlated representation. Use the coupling modeling method to model the performance influence between different systems in a multi-level and multi-scale environment to reveal the relationship between decomposition and coupling between the agricultural system and the energy system;
[0049] A time scale decomposition model for decomposing agricultural systems and energy systems is established. The fast system in the time scale decomposition model is defined to include environmental parameters and consumption parameters. Environmental parameters include temperature T, humidity H, and light intensity I. Consumption parameters include power consumption P. f 、Energy load L f The slow system in the time scale decomposition model includes growth parameters and allocation parameters. Growth parameters include growth stage G(t) and land nutrients N, and allocation parameters are energy reserves E. s ;
[0050] The fast system equation is established as Where, X f is the parameter variable of the fast system, and the equation of the slow system is established as Where, X s is the parameter variable of the slow system, and the coupling term is set to G(X f ,X s ) represents the linkage relationship between the fast system and the slow system. The fast system after linkage is The slow system after linkage is
[0051] It should be noted that the fast system is a short-term variable, including environmental control factors in facility agriculture, such as light, temperature, humidity, etc. Environmental control factors change rapidly in daily operations and require timely monitoring and adjustment. The slow system is a long-term variable, including energy storage and distribution, agricultural production cycles, including crop growth cycles, soil nutrient changes, etc. Long-term variables change slowly and have a longer time span, and the time span unit is usually years.
[0052] The method of obtaining sensitive error information based on the time scale decomposition model is:
[0053] Obtain the sensitivity response coefficients of the time scale decomposition model within the periodic time Q to the input parameters of the fast system and the slow system, and integrate the sensitivity response coefficients of the time scale decomposition model within several periodic time Q to the input parameters of the fast system and the slow system into a short-term set and a long-term set, marking the short-term set as Ssy = {sr p}, where p = {1, 2, 3, ..., o}, and o is a positive integer, sr p It represents the sensitivity response coefficient of the time scale decomposition model within the pth period time Q to the slow system input parameter. The calculation expression of the sensitivity response coefficient is: Where T represents the time scale, and the long-term set is Lsy={lr u}, where u = {1, 2, 3, ..., y}, and y is a positive integer, lr u It represents the sensitivity response coefficient of the time scale decomposition model within the u-th cycle time Q to the slow system input parameters. The calculation expression of the sensitivity response coefficient is: Where G represents the coupling term;
[0054] Calculate the average standard deviation of the sensitivity response coefficient of the time scale decomposition model to the input parameters of the fast system and the slow system within several cycle times Q. The calculation expression is: Where avg1 is the average value of the sensitivity response coefficient of the time scale decomposition model to the input parameters of the fast system within several cycle times Q. The calculation expression is: avg2 is the average value of the sensitivity response coefficient of the time scale decomposition model to the slow system input parameters within several cycle times Q. The calculation expression is:
[0055] By obtaining the sensitive error information of the time scale decomposition model, the fast system and the slow system are equally decomposed. The influence of the input variables on the output results is evaluated under the premise of avoiding time scale interference, and the underestimation or overestimation of the influence of the input variables in the time scale division process is avoided, thereby optimizing the overall accuracy of the time scale division. At the same time, it is beneficial to evaluate the influence of variables on the coupling between the fast system and the slow system of the system, and to adjust the time scale division in time.
[0056] The method for obtaining the decomposition difference coefficient through sensitive error information is:
[0057] The gradient descent algorithm is used to adjust the weight coefficient of the coupling term, the weight coefficient of the coupling term is marked as θ, and the coupling term is defined as G(X f ,X s ,θ)=θ·f(X f ,X s ), where f is the coupling function between the fast system and the slow system, and the difference between the output value after the coupling and the actual value is calibrated In the formula, n is the number of samples, i is the sample index, is the output value after coupling, Y i is the actual value, L(θ) is the loss function, and the gradient of the loss function to the weight is calculated by derivation. The calculation expression is: In the formula, the difference between the weight update and the weight update is calculated by gradient descent, and the expression is In the formula, α is the learning rate, and the calculation expression of the decomposition difference coefficient is Where D dc is the coefficient of decomposition.
[0058] The gradient descent algorithm is an optimization algorithm used to adjust the weight coefficient of the coupling term. By analyzing the sensitivity of the input variables, the sensitivity of the system to different input parameters is evaluated, and the influence of different variables on the model results is identified, thereby optimizing the accuracy of the time scale division.
[0059] The decomposition difference coefficient is obtained through sensitive error information, and the results after coupling the fast system and the slow system are compared with the actual values to calculate the actual modeling error, evaluate the effectiveness of the time scale decomposition model, avoid the situation where the model fails due to time scale coordination, optimize the time coupling relationship between the fast system and the slow system, and further improve the accuracy of the model. The coupling of the weight coefficients of the coupling terms is calculated by the gradient descent method to capture the dynamic interaction between the fast system and the slow system, and optimize the interaction mode of the fast system and the slow system in the model. The coupling effect of the fast system and the slow system on the time scale division is verified in the dynamic change process of the fast system and the slow system, avoid the system prediction error caused by improper setting of the coupling parameters, and solve the dynamic adjustment problem in the coupling model.
[0060] According to the risk linkage relationship of energy security coordination, coupling fluctuation information is obtained, and the coordination load coefficient is obtained according to the coupling fluctuation information;
[0061] The method for obtaining coupling fluctuation information is:
[0062] Obtain the abnormal response coefficient of the data interaction between the agricultural system and the energy system within the cycle time T, and integrate the abnormal response coefficients of the data interaction between the agricultural system and the energy system within several cycle times T into a data set, and mark the data set as E xr ={in d}, where d = {1, 2, 3, ..., f}, and f is a positive integer, in d It represents the data interaction abnormal response coefficient between the agricultural system and the energy system within the dth cycle time T. The calculation expression of the data interaction abnormal response coefficient is: Among them, t h is the energy warning response time, t j The energy warning will stop reporting time. ae Data transmitted from the agricultural system to the energy system, D ea data transmitted from energy systems to agricultural systems;
[0063] Calculate the standard deviation of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T. The calculation expression is: Where res is the average value of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T, and the calculation expression is:
[0064] The coupling fluctuation information is obtained through the data interaction abnormal response coefficient to evaluate the coupling state of the energy system and the agricultural system. When the nonlinear state of the coupling relationship produces fluctuation abnormalities, the update frequency of the state information between the energy system and the agricultural system increases. The coupling fluctuation information effectively represents the fluctuation state of the data interaction between the energy system and the agricultural system.
[0065] The method for obtaining the cooperative load factor based on the coupled fluctuation information is:
[0066] The system equation for calibrating the coupling relationship between the agricultural system and the energy system is Z, which is expressed as Among them, X f is the input variable of the agricultural system, X s is the input variable of the energy system, F(X f , X s ) is the nonlinear coupling equation of the agricultural system, S(X f , X s ) is the nonlinear coupling equation of the energy system. The numerical integration method is used to calculate the Lyapunov index of the system equation Z. The calculation expression is: Among them, Z 0 is the initial state of the system equation, δ is a very small constant, indicating a state with a small difference from the initial state, and δ is a positive number, t is the time, then the calculation expression of the collaborative load coefficient is C of =C om λ+1 .
[0067] By calculating the Lyapunov exponent of the system equation composed of the coupling relationship between the agricultural system and the energy system, the nonlinear fluctuation degree of the integration of the energy system and the agricultural system is tested. Under the coordinated early warning state of abnormal environmental factors, imbalance in energy supply and demand, and abnormal agricultural production, the nonlinear degree of the coupled system increases sharply, that is, the dependence between the energy system and the agricultural system makes the abnormal fluctuation of one system affect the other system. There are potential risks in the stable operation of the coupled system. The system reacts violently to disturbances and has significant nonlinear effects. For example, slight environmental changes may cause large fluctuations in the production efficiency of the agricultural system, thereby affecting the energy consumption state. By calculating the coordinated load coefficient, the anti-disturbance state of the coupled system is effectively revealed, and the nonlinear state of the coupled system is detected, which provides strong support for the modeling credibility model based on the coupled system, thereby verifying the rationality of the coordinated early warning judgment logic.
[0068] The modeling confidence model is constructed based on the decomposition difference coefficient and the collaborative load coefficient. The modeling robustness of the energy security collaborative early warning is evaluated by the ability of the coupling term to affect the fluctuation of the time scale decomposition model, and the model robustness index is generated.
[0069] The logic of constructing the modeling reliability model based on the decomposition difference coefficient and the collaborative load coefficient is:
[0070] The expression of the modeling confidence model is: Where M re is the model robustness index, F reis the frequency of data transmission between the energy system and the agricultural system within the cycle time T.
[0071] The modeling credibility model obtains a verification model for modeling and analysis of nonlinear coupling between energy and agricultural systems by comprehensively evaluating the decomposition difference coefficient and the collaborative load coefficient. It effectively refines the ability to analyze the linkage effects of nonlinear fluctuations and fluctuations on energy and agricultural systems, and verifies the effectiveness of early warning collaborative analysis.
[0072] The model performance is graded according to the time cost and computing cost of the modeling credibility model analysis, a baseline robustness threshold is preset, and the model robustness index is compared with the baseline robustness threshold, and the stability of the modeling credibility model is classified according to the comparison results;
[0073] The logic of grading model performance based on the time cost and computing power cost of modeling credibility analysis is as follows:
[0074] The computation time for calibrating the modeling confidence model is C pt , the energy consumption of the modeling credibility model is E co , the model performance index is calculated as Preset the first performance threshold M f and the second performance threshold M s , and the first performance threshold M f Less than the second performance threshold M s , when the calculated model performance index M dp Less than or equal to the first performance threshold M f When the model performance is defined as M1, the calculated first performance threshold M f Less than the model performance index M dp Less than the second performance threshold M s When the model performance is defined as M2, the calculated model performance index M dp Greater than or equal to the second performance threshold M s When , the model performance is defined as M3 level;
[0075] The logic for classifying the stability of the modeling confidence model is:
[0076] Preset benchmark robust threshold B as , the model robustness index M re With the benchmark robust threshold B as When the calculated model robustness index M re Greater than or equal to the benchmark robust threshold B as When , the modeling confidence model is marked as sensitive. When the calculated model robustness index M re Less than the benchmark robustness threshold B as When , the marker modeling reliability model is hysteresis type;
[0077] Generate early warning verification intervention strategies based on the grading and classification results for the effectiveness of collaborative early warning and inform management personnel;
[0078] When the model performance of the modeling confidence model is at level M1 and the model type is sensitive, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is large, the short-term warning credibility is high, and the long-term warning credibility is low;
[0079] When the model performance of the modeling confidence model is at level M2 or M3, and the model type is sensitive, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is large, the short-term warning credibility is medium, and the long-term warning credibility is low;
[0080] When the model performance of the modeling confidence model is at level M1 and the model type is hysteresis type, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is small, the short-term warning credibility is high, and the long-term warning credibility is high;
[0081] When the model performance of the modeling confidence model is at level M2 or M3, and the model type is hysteresis type, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is small, the short-term warning credibility is low, and the long-term warning credibility is medium.
[0082] This application decomposes the time scale of the agricultural system and the energy system by establishing a time scale decomposition model, and obtains sensitive error information, evaluates the influence of input variables on output results without being disturbed by the time scale, avoids the situation where the influence of input variables is underestimated or overestimated during the time scale division process, and thus optimizes the overall accuracy of the time scale division. By obtaining the decomposition difference coefficient, the coupling system based on the energy system and the agricultural system is evaluated for its ability to resist disturbance inputs, and the correlation analysis effect of the coupling system is tested. The decomposition difference coefficient and the collaborative load coefficient are comprehensively evaluated through the modeling credibility model, and a verification model for the modeling analysis of the nonlinear coupling of the energy system and the agricultural system is obtained. The ability to effectively refine the linkage influence of nonlinear fluctuations and fluctuations on the energy system and the agricultural system is verified, which provides verification for the effectiveness of the early warning collaborative analysis. By decomposing from the perspective of fast and slow systems and then coupling from the perspective of energy and agriculture, the difference between the decomposition scale and the coupling scale is effectively resolved, and the degree of fusion of the coupling system is further improved, thereby optimizing the accuracy of the correlation risk analysis and the effectiveness of the collaborative early warning.
[0083] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0084] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of computer program goods. The computer program goods include one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0085] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0086] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0087] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0088] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A collaborative early warning method for energy security in an agricultural park, characterized in that: The specific steps include: Obtain sensitive error information based on the time scale decomposition model used to decompose the agricultural system and the energy system, and obtain the decomposition difference coefficient based on the sensitive error information; According to the risk linkage relationship of energy security coordination, coupling fluctuation information is obtained, and the coordination load coefficient is obtained according to the coupling fluctuation information; The modeling confidence model is constructed based on the decomposition difference coefficient and the collaborative load coefficient. The modeling robustness of the energy security collaborative early warning is evaluated by the ability of the coupling term to affect the fluctuation of the time scale decomposition model, and the model robustness index is generated. The model performance is graded according to the time cost and computing cost of the modeling credibility model analysis, a baseline robustness threshold is preset, and the model robustness index is compared with the baseline robustness threshold, and the stability of the modeling credibility model is classified according to the comparison results; Based on the grading and classification results, an early warning verification intervention strategy is generated for the effectiveness of the collaborative early warning and notified to the management personnel.
2. The energy security collaborative early warning method for an agricultural park according to claim 1 is characterized in that: The time scale decomposition model for decomposing agricultural and energy systems is constructed as follows: A time scale decomposition model for decomposing agricultural systems and energy systems is established. The fast system in the time scale decomposition model is defined to include environmental parameters and consumption parameters. Environmental parameters include temperature T, humidity H, and light intensity I. Consumption parameters include power consumption P. f 、Energy load L f The slow system in the time scale decomposition model includes growth parameters and allocation parameters. Growth parameters include growth stage G(t) and land nutrients N, and allocation parameters are energy reserves E. s ; The fast system equation is established as Where, X f is the parameter variable of the fast system, and the equation of the slow system is established as Where, X s is the parameter variable of the slow system, and the coupling term is set to G(X f ,X s ) represents the linkage relationship between the fast system and the slow system. The fast system after linkage is The slow system after linkage is 3. The energy security collaborative early warning method for an agricultural park according to claim 2 is characterized in that: The method of obtaining sensitive error information based on the time scale decomposition model is: Obtain the sensitivity response coefficients of the time scale decomposition model within the periodic time Q to the input parameters of the fast system and the slow system, and integrate the sensitivity response coefficients of the time scale decomposition model within several periodic time Q to the input parameters of the fast system and the slow system into a short-term set and a long-term set, marking the short-term set as Ssy = {sr p }, where p = {1, 2, 3, ..., o}, and o is a positive integer, sr p It represents the sensitivity response coefficient of the time scale decomposition model within the pth period time Q to the slow system input parameter. The calculation expression of the sensitivity response coefficient is: Where T represents the time scale, and the long-term set is Lsy={lr u }, where u = {1, 2, 3, ..., y}, and y is a positive integer, lr u It represents the sensitivity response coefficient of the time scale decomposition model within the u-th cycle time Q to the slow system input parameters. The calculation expression of the sensitivity response coefficient is: Where G represents the coupling term; Calculate the average standard deviation of the sensitivity response coefficient of the time scale decomposition model to the input parameters of the fast system and the slow system within several cycle times Q. The calculation expression is: Where avg1 is the average value of the sensitivity response coefficient of the time scale decomposition model to the fast system input parameters within several cycle times Q. The calculation expression is: avg2 is the average value of the sensitivity response coefficient of the time scale decomposition model to the slow system input parameters within several cycle times Q. The calculation expression is:
4. The energy security collaborative early warning method for an agricultural park according to claim 3 is characterized in that: The method for obtaining the decomposition difference coefficient through sensitive error information is: The gradient descent algorithm is used to adjust the weight coefficient of the coupling term, the weight coefficient of the coupling term is marked as θ, and the coupling term is defined as G(X f ,X s ,θ)=θ·f(X f ,X s ), where f is the coupling function between the fast system and the slow system, and the difference between the output value after the coupling and the actual value is calibrated In the formula, n is the number of samples, i is the sample index, is the output value after coupling, Y i is the actual value, L(θ) is the loss function, and the gradient of the loss function to the weight is calculated by derivation. The calculation expression is: In the formula, the difference between the weight update and the weight update is calculated by gradient descent, and the expression is In the formula, α is the learning rate, and the calculation expression of the decomposition difference coefficient is Where D dc is the coefficient of decomposition.
5. The energy security collaborative early warning method for an agricultural park according to claim 1 is characterized in that: The method for obtaining coupling fluctuation information is: Obtain the abnormal response coefficient of the data interaction between the agricultural system and the energy system within the cycle time T, and integrate the abnormal response coefficients of the data interaction between the agricultural system and the energy system within several cycle times T into a data set, and mark the data set as E xr ={in d }, where d = {1, 2, 3, ..., f}, and f is a positive integer, in d It represents the data interaction abnormal response coefficient between the agricultural system and the energy system within the dth cycle time T. The calculation expression of the data interaction abnormal response coefficient is: Among them, t h is the energy warning response time, t j The energy warning will stop reporting time. ae Data transmitted from the agricultural system to the energy system, D ea data transmitted from energy systems to agricultural systems; Calculate the standard deviation of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T. The calculation expression is: Where res is the average value of the abnormal response coefficient of the data interaction between the agricultural system and the energy system within several cycles T, and the calculation expression is:
6. The energy security collaborative early warning method for an agricultural park according to claim 5 is characterized in that: The method for obtaining the cooperative load factor based on the coupled fluctuation information is: The system equation for calibrating the coupling relationship between the agricultural system and the energy system is Z, which is expressed as Among them, X f is the input variable of the agricultural system, X s is the input variable of the energy system, F(X f , X s ) is the nonlinear coupling equation of the agricultural system, S(X f , X s ) is the nonlinear coupling equation of the energy system. The numerical integration method is used to calculate the Lyapunov index of the system equation Z. The calculation expression is: Where Z0 is the initial state of the system equation, δ is a very small constant, indicating a state with a small difference from the initial state, and δ is a positive number, t is time, then the calculation expression of the collaborative load coefficient is C of =C om λ+1 .
7. The energy security collaborative early warning method for an agricultural park according to claim 6 is characterized in that: The logic of constructing the modeling reliability model based on the decomposition difference coefficient and the collaborative load coefficient is: The expression of the modeling confidence model is: Where M re is the model robustness index, F re is the frequency of data transmission between the energy system and the agricultural system within the cycle time T.
8. The energy security collaborative early warning method for an agricultural park according to claim 7 is characterized in that: The logic of grading model performance based on the time cost and computing power cost of modeling credibility analysis is as follows: The computation time for calibrating the modeling confidence model is C pt , the energy consumption of the modeling credibility model is E co , the model performance index is calculated as Preset the first performance threshold M f and the second performance threshold M s , and the first performance threshold M f Less than the second performance threshold M s , when the calculated model performance index M dp Less than or equal to the first performance threshold M f When the model performance is defined as M1, the calculated first performance threshold M f Less than the model performance index M dp Less than the second performance threshold M s When the model performance is defined as M2, the calculated model performance index M dp Greater than or equal to the second performance threshold M s When , the model performance is defined as M3 level.
9. The energy security collaborative early warning method for an agricultural park according to claim 8, characterized in that: The logic for classifying the stability of the modeling confidence model is: Preset benchmark robust threshold B as , the model robustness index M re With the benchmark robust threshold B as When the calculated model robustness index M re Greater than or equal to the benchmark robust threshold B as When , the modeling confidence model is marked as sensitive. When the calculated model robustness index M re Less than the benchmark robustness threshold B as When , the marker modeling reliability model is hysteresis type.
10. The energy security collaborative early warning method for an agricultural park according to claim 9, characterized in that: The logic of generating early warning verification intervention strategies based on the effectiveness of collaborative early warnings based on the grading and classification results and informing management personnel is as follows: When the model performance of the modeling confidence model is at level M1 and the model type is sensitive, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is large, the short-term warning credibility is high, and the long-term warning credibility is low; When the model performance of the modeling confidence model is at level M2 or M3, and the model type is sensitive, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is large, the short-term warning credibility is medium, and the long-term warning credibility is low; When the model performance of the modeling confidence model is at level M1 and the model type is hysteresis type, it prompts managers that the coupling accuracy is high and the nonlinear fluctuation is small, the short-term warning credibility is high, and the long-term warning credibility is high; When the model performance of the modeling confidence model is at level M2 or M3, and the model type is hysteresis type, it prompts managers that the coupling accuracy is low and the nonlinear fluctuation is small, the short-term warning credibility is low, and the long-term warning credibility is medium.