Energy control method and system of wireless sensor network

Abstract

The invention provides an energy control method and system for a wireless sensor network. The method comprises the following steps: acquiring working states and state conversion time of a plurality of wireless sensors of the wireless sensor network; inputting the working state and the state conversion time into a pre-constructed energy consumption prediction model to obtain an energy consumption prediction result output by the energy consumption prediction model; according to the energy consumption prediction result, determining a data quality demand, evaluating a data acquisition demand, and obtaining a work demand evaluation result; based on the work demand evaluation result, dynamically adjusting the sleep periods and data transmission parameters of the plurality of wireless sensors; and monitoring the adjusted energy consumption levels of the plurality of wireless sensors, identifying abnormal nodes, and adjusting the connection paths of the plurality of wireless sensors. According to the invention, energy consumption of the sensor can be controlled in real time, abnormal nodes can be identified in time, network connection is optimized, and operation efficiency, stability and reliability of the sensor network are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless sensor networks, and in particular to an energy control method and system for a wireless sensor network. Background Art

[0002] The energy control method of wireless sensor networks focuses on optimizing energy usage in wireless sensor networks, aiming to extend the operating life of the sensor network and improve the overall efficiency, ensure continuous data collection and transmission in an energy-constrained environment, and ensure the performance of wireless sensors and reduce ineffective energy consumption through energy control.

[0003] The energy control method of traditional wireless sensor networks uses a static control strategy based on preset conditions. In practical applications, it is difficult to adapt to rapidly changing needs, resulting in insufficient energy use and delayed response, and in emergencies and abnormal working conditions, leading to waste of resources and insufficient utilization. There are deficiencies in real-time fault diagnosis and early warning, and a lack of timely and effective anomaly detection methods, which increases the difficulty and cost of system operation and maintenance, and affects the operating efficiency, stability and reliability of the sensor network. Summary of the invention

[0004] The present invention provides an energy control method and system for a wireless sensor network, which are used to solve the problems in the prior art of using a static control strategy to control the energy of a wireless sensor network, which has poor real-time performance, is difficult to adapt to rapidly changing complex application scenarios, and has deficiencies in real-time fault diagnosis and early warning, thus affecting the operating efficiency, stability and reliability of the sensor network.

[0005] In a first aspect, the present invention provides an energy control method for a wireless sensor network, comprising: Obtain the working status and state transition time of multiple wireless sensors in a wireless sensor network; Inputting the working states and state transition times of the multiple wireless sensors into a pre-built energy consumption prediction model to obtain energy consumption prediction results of the multiple wireless sensors output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on multiple historical working states, multiple historical state transition times and historical energy consumption analysis results of multiple sample wireless sensors; Determine data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluate data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result; Based on the work demand evaluation result, the sleep cycles of the multiple wireless sensors are adjusted to obtain sleep parameters of the multiple wireless sensors, and based on the sleep parameters, the data transmission parameters of the multiple wireless sensors are adjusted; The adjusted energy consumption levels of the multiple wireless sensors are monitored, wireless sensors with abnormal energy consumption are identified, abnormal node detection records are generated, and connection paths of the multiple wireless sensors are adjusted based on the abnormal node detection records to obtain control parameters of the wireless sensor network.

[0006] In some embodiments, adjusting the connection paths of the plurality of wireless sensors based on the abnormal node detection record includes: Based on the abnormal node detection record, the connection data and load information of the multiple wireless sensors are monitored in real time, the connection efficiency and the overall network load between the multiple wireless sensors are evaluated, and the real-time network load information is generated; Based on the real-time network load information and the working states of the multiple wireless sensors, and the physical distances between the multiple wireless sensors, the Dijkstra algorithm is used to calculate the optimal paths of the multiple wireless sensors, adjust the connection paths of the multiple wireless sensors, and generate connection path adjustment records; The transmission performance data of the wireless sensor network is monitored in real time, and abnormal wireless sensors with connection bottlenecks are identified according to the connection path adjustment records and the transmission performance data, and the connection protocol type of the abnormal wireless sensors is adjusted.

[0007] In some embodiments, the calculation formula of the optimal path of the multiple wireless sensors is as follows: ; in, Represents the distance from the starting wireless sensor node to the target wireless sensor node The shortest path length, Represents the distance from the starting wireless sensor node to the intermediate wireless sensor node The shortest path length, For Node To Node The basic weight of For communication quality The corresponding adjustment factor is For Node and nodes The quality of communication between For the current load The corresponding adjustment factor is For Node The current load, For the current load The corresponding adjustment factor is For Node The current load, Logical distance The corresponding adjustment factor is For Node and nodes The logical distance between them.

[0008] In some embodiments, monitoring the adjusted energy consumption levels of the plurality of wireless sensors, identifying wireless sensors with abnormal energy consumption, and generating abnormal node detection records includes: Collecting the adjusted energy consumption data of the multiple wireless sensors, evaluating the energy consumption change patterns and energy consumption fluctuation ranges of the multiple wireless sensors, and generating real-time energy consumption monitoring data; Based on the real-time energy consumption monitoring data, identifying wireless sensors with abnormal energy consumption and generating abnormal node information; Generate warning information based on the abnormal node information, send the warning information to the user end, and obtain the abnormal node detection record.

[0009] In some embodiments, adjusting the sleep cycles of the multiple wireless sensors based on the work demand evaluation result to obtain the sleep parameters of the multiple wireless sensors includes: Based on the work demand evaluation result, determine the data collection frequency requirement of each wireless sensor, based on the current energy consumption and data collection frequency requirement of each wireless sensor, determine the optimal working time period required by each wireless sensor, and generate working time period information; Based on the working period information, evaluating the impact of various sleep cycles on the performance of the wireless sensor network, and generating sleep parameter evaluation data; Based on the sleep parameter evaluation data, the sleep parameters of the wireless sensor network are calculated and adjusted in consideration of the energy consumption and data quality requirements of the multiple wireless sensors, so as to determine the optimal sleep parameters of the wireless sensor network.

[0010] In some embodiments, adjusting the data transmission parameters of the plurality of wireless sensors based on the sleep parameters includes: Based on the sleep parameters, control the target wireless sensor to sleep, evaluate the transmission performance of the wireless sensor network in real time, and generate network load analysis data; Based on the network load analysis data and the data quality requirements, calculating optimal data transmission parameters of the plurality of wireless sensors, the optimal data transmission parameters comprising an optimal data packet capacity and an optimal data transmission frequency; Based on the optimal data transmission parameters, the data transmission frequency and data packet capacity of the multiple wireless sensors are adjusted, the communication efficiency and response speed of the wireless sensor network are monitored in real time, and the transmission parameter adjustment result is obtained.

[0011] In some embodiments, determining the data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluating the data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain the work demand evaluation results includes: Acquire geographic locations and network performance indicators of the multiple wireless sensors; Determine data quality requirements of the multiple wireless sensors according to energy consumption prediction results of the multiple wireless sensors, as well as geographic locations and network performance indicators of the multiple wireless sensors, wherein the data quality requirements include real-time requirements and accuracy requirements; estimating power consumption levels of the plurality of wireless sensors at differentiated data collection frequencies, and generating power consumption level analysis results; Based on the power consumption level analysis results and data quality requirements, optimal data collection frequency parameters of the multiple wireless sensors are calculated to obtain work requirement evaluation results of the multiple wireless sensors.

[0012] In some embodiments, the training process of the energy consumption prediction model includes: Acquire historical operation log data of multiple sample wireless sensors of the sample wireless sensor network, process the historical operation log data, obtain multiple historical working states and multiple historical state conversion times of the multiple sample wireless sensors, and historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Determine historical energy consumption analysis results of the multiple sample wireless sensors based on historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Inputting a plurality of historical working states and a plurality of historical state conversion times of the plurality of sample wireless sensors into an initial energy consumption prediction model to obtain energy consumption prediction data of the plurality of sample wireless sensors output by the initial energy consumption prediction model; Calculating a loss function value based on the energy consumption prediction data of the plurality of sample wireless sensors and the historical energy consumption analysis results of the plurality of sample wireless sensors; Based on the loss function value, the parameters of the initial energy consumption prediction model are iteratively optimized to obtain the energy consumption prediction model.

[0013] In some embodiments, determining the historical energy consumption analysis results of the plurality of sample wireless sensors based on the historical energy consumption data of the plurality of sample wireless sensors under a plurality of historical working states includes: Calculate the historical average energy consumption of each sample wireless sensor based on the historical energy consumption data of each sample wireless sensor under multiple historical working states; Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical energy consumption level and the historical energy consumption fluctuation range of each sample wireless sensor under various historical working conditions are calculated to obtain the historical energy consumption comparison information of each sample wireless sensor; Based on the historical energy consumption comparison information of each sample wireless sensor, the impact of different historical working states and different historical state conversion conditions on the historical energy consumption of each sample wireless sensor is evaluated, and combined with the historical average energy consumption of each sample wireless sensor, the historical energy consumption analysis result of each sample wireless sensor is obtained.

[0014] In a second aspect, the present invention further provides an energy control system for a wireless sensor network, comprising: An acquisition unit, used for acquiring the working status and state transition time of multiple wireless sensors of the wireless sensor network; A prediction unit, configured to input the working states and state transition times of the plurality of wireless sensors into a pre-built energy consumption prediction model, and obtain energy consumption prediction results of the plurality of wireless sensors output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on a plurality of historical working states, a plurality of historical state transition times, and historical energy consumption analysis results of a plurality of sample wireless sensors; An evaluation unit, configured to determine data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluate data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result; An adjusting unit, configured to adjust the sleep cycles of the plurality of wireless sensors based on the work demand evaluation result, obtain sleep parameters of the plurality of wireless sensors, and adjust the data transmission parameters of the plurality of wireless sensors based on the sleep parameters; The monitoring unit is used to monitor the adjusted energy consumption levels of the multiple wireless sensors, identify wireless sensors with abnormal energy consumption, generate abnormal node detection records, and adjust the connection paths of the multiple wireless sensors based on the abnormal node detection records.

[0015] The energy control method and system of the wireless sensor network provided by the present invention obtains the working status and state transition time of multiple wireless sensors of the wireless sensor network, inputs the working status and state transition time into a pre-built energy consumption prediction model, obtains the energy consumption prediction result output by the energy consumption prediction model, determines the data quality requirements according to the energy consumption prediction result, evaluates the data collection requirements, obtains the work demand evaluation result, and dynamically adjusts the sleep cycle and data transmission parameters of multiple wireless sensors based on the work demand evaluation result, monitors the adjusted energy consumption levels of multiple wireless sensors, identifies abnormal nodes, and adjusts the connection paths of multiple wireless sensors. The present invention can control the energy consumption of sensors in real time, identify abnormal nodes in time, optimize network connections, and improve the operating efficiency, stability and reliability of the sensor network. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of an energy control method for a wireless sensor network provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a process of adjusting the sleep cycles of multiple wireless sensors based on work demand evaluation results provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of a process of adjusting connection paths of multiple wireless sensors based on abnormal node detection records provided by an embodiment of the present invention; Figure 4 It is a flowchart of the training process of the energy consumption prediction model provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of the structure of an energy control system of a wireless sensor network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0019] Figure 1A schematic diagram of a flow chart of an energy control method for a wireless sensor network provided by an embodiment of the present invention. Figure 1 As shown, a method for energy control of a wireless sensor network is provided, comprising the following steps: step 110, step 120, step 130, step 140 and step 150. The steps of the method flow are only a possible implementation of the present invention.

[0020] Step 110: Acquire the working states and state transition times of multiple wireless sensors of the wireless sensor network.

[0021] Optionally, operation log data of multiple wireless sensors of the wireless sensor network is acquired, and the operation log data is processed to obtain the working states and state transition times of the multiple wireless sensors.

[0022] Optionally, the operation log data is preprocessed by data cleaning, denoising, standardization, etc. to obtain preprocessed operation log data.

[0023] Optionally, feature extraction is performed on the preprocessed operation log data to obtain the working states and state transition times of the multiple wireless sensors.

[0024] Step 120: Input the working states and state transition times of the multiple wireless sensors into a pre-built energy consumption prediction model to obtain energy consumption prediction results of the multiple wireless sensors output by the energy consumption prediction model. The energy consumption prediction model is trained based on multiple historical working states, multiple historical state transition times and historical energy consumption analysis results of multiple sample wireless sensors.

[0025] Optionally, the energy consumption prediction result includes at least average energy consumption, energy consumption fluctuation range, and energy consumption peak data under multiple working states.

[0026] Optionally, the calculation formula of the energy consumption prediction model is as follows: ; in, is the predicted value of energy consumption of wireless sensors, is the working status of the wireless sensor, is the state transition time of the wireless sensor, is the intercept, is the coefficient of the working state affecting energy consumption, is the coefficient of the state transition time on energy consumption.

[0027] It can be understood that by adopting a pre-built energy consumption prediction model, energy consumption prediction can be quickly performed based on the working status and state transition time of multiple wireless sensors, and energy consumption prediction results can be obtained, providing a data basis for subsequent energy control, which helps to improve the real-time performance of sensor energy control.

[0028] Step 130: determine data quality requirements of the multiple wireless sensors based on the energy consumption prediction results of the multiple wireless sensors, and evaluate data collection requirements of the multiple wireless sensors based on the data quality requirements to obtain work requirement evaluation results.

[0029] Optionally, the data quality requirements include real-time requirements and accuracy requirements.

[0030] Optionally, the work demand assessment result includes at least data collection frequency parameters, the amount of collected data, and the data update frequency.

[0031] It is understandable that by determining the data quality requirements of multiple wireless sensors based on the energy consumption prediction results of multiple wireless sensors, it is possible to reduce energy consumption while ensuring the timeliness and accuracy of data and reducing the risk of node overload.

[0032] Step 140: Based on the work demand evaluation result, the sleep cycles of the multiple wireless sensors are adjusted to obtain sleep parameters of the multiple wireless sensors, and based on the sleep parameters, the data transmission parameters of the multiple wireless sensors are adjusted.

[0033] Optionally, the sleep parameters include at least a node sleep time, a node working time, and a sleep threshold parameter.

[0034] Optionally, the data transmission parameters include at least data transmission frequency, data packet capacity, and transmission priority parameters.

[0035] It can be understood that by adjusting the sleep cycles and transmission frequencies of multiple wireless sensors, the response speed and energy consumption level of the wireless sensor network can be balanced.

[0036] Step 150: monitor the adjusted energy consumption levels of the multiple wireless sensors, identify wireless sensors with abnormal energy consumption, generate abnormal node detection records, and adjust connection paths of the multiple wireless sensors based on the abnormal node detection records.

[0037] Optionally, the abnormal node detection record includes at least a list of abnormal energy consumption nodes, the time when the abnormality occurred, and a warning information sending record.

[0038] Optionally, based on the abnormal node detection record, the connection paths of the plurality of wireless sensors are adjusted to obtain control parameters of the wireless sensor network.

[0039] Optionally, the control parameters include at least a wireless sensor network topology diagram, a node connection path, and a node connection protocol type.

[0040] It can be understood that by monitoring the adjusted energy consumption levels of multiple wireless sensors, identifying wireless sensors with abnormal energy consumption, generating abnormal node detection records, and adjusting the connection paths of multiple wireless sensors based on the abnormal node detection records, the stability and reliability of the sensor network can be improved, and the stability and delay of network data transmission can be optimized.

[0041] In an embodiment of the present invention, by acquiring operation log data of a wireless sensor network, the working status and state transition time of a plurality of wireless sensors are obtained, the working status and state transition time are input into a pre-built energy consumption prediction model, and an energy consumption prediction result output by the energy consumption prediction model is obtained. According to the energy consumption prediction result, data quality requirements are determined, data collection requirements are evaluated, and work demand evaluation results are obtained. Based on the work demand evaluation results, the sleep cycles and data transmission parameters of the plurality of wireless sensors are dynamically adjusted, the adjusted energy consumption levels of the plurality of wireless sensors are monitored, abnormal nodes are identified, and the connection paths of the plurality of wireless sensors are adjusted. The energy consumption of the sensors can be controlled in real time, abnormal nodes can be identified in time, network connections can be optimized, and the operation efficiency, stability and reliability of the sensor network can be improved.

[0042] In some embodiments, step 130, determining data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluating data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result, includes: Step 131: Obtaining geographic locations and network performance indicators of multiple wireless sensors; Step 132: determining data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, as well as the geographical locations and network performance indicators of the multiple wireless sensors, where the data quality requirements include real-time requirements and accuracy requirements; Step 133: estimating the power consumption levels of the plurality of wireless sensors at the differentiated data collection frequencies, and generating a power consumption level analysis result; Step 134: Based on the power consumption level analysis results and the data quality requirements, the optimal data collection frequency parameters of the multiple wireless sensors are calculated to obtain the work requirement evaluation results of the multiple wireless sensors.

[0043] Optionally, data quality requirements and overall network performance are evaluated based on energy consumption prediction results of the multiple wireless sensors, as well as geographic locations and network performance indicators of the multiple wireless sensors.

[0044] Optionally, the network performance score is calculated as follows: ; in, Represents the network performance score of the wireless sensor network, For wireless sensor nodes The real-time data requirements For wireless sensor nodes The relative contribution of geographic location to network performance, and is the weight coefficient, is the total number of nodes.

[0045] Optionally, based on the data real-time requirements and the impact of the geographical location, the impact of the node location on the network performance is quantified to generate node location requirement data.

[0046] Optionally, based on the node location demand data, the power consumption levels of the wireless sensors at various data collection frequencies are analyzed to generate power consumption level analysis results.

[0047] Optionally, a power consumption model corresponding to the data acquisition frequency is pre-constructed, and the power consumption model is used to calculate the average power consumption and fluctuation range at each data acquisition frequency.

[0048] Optionally, the average power consumption is calculated as follows: ; in, is the average power consumption of the wireless sensor, The data collection frequency The power consumption of wireless sensors is is the power consumption adjustment coefficient, which reflects the variation of power consumption with acquisition frequency. The number of categories for which data is collected for evaluation.

[0049] Optionally, the average data collection frequency required by the wireless sensor network is calculated using the following formula: ; in, is the average data collection frequency required by the wireless sensor network, For Node The average power consumption, For Node data quality requirements, and is the weight coefficient, is the total number of nodes.

[0050] It should be noted that, combined with the average power consumption and data quality requirements, the optimal data collection frequency of multiple wireless sensor nodes is calculated by weighted average to ensure the balance between energy efficiency and data quality and obtain the work demand assessment result.

[0051] Figure 2The present invention provides a flowchart of adjusting the sleep cycles of multiple wireless sensors based on the work demand evaluation results. Figure 2 As shown, in some embodiments, in step 140, based on the work demand evaluation result, the sleep cycles of the multiple wireless sensors are adjusted to obtain the sleep parameters of the multiple wireless sensors, including: Step 141: based on the work demand evaluation result, determine the data collection frequency requirement of each wireless sensor, based on the current energy consumption and data collection frequency requirement of each wireless sensor, determine the optimal working time period required for each wireless sensor, and generate working time period information; Step 142: Based on the working period information, evaluate the impact of various sleep cycles on the performance of the wireless sensor network and generate sleep parameter evaluation data; Step 143: Based on the sleep parameter evaluation data, considering the energy consumption and data quality requirements of multiple wireless sensors, calculate and adjust the sleep parameters of the wireless sensor network, and determine the optimal sleep parameters of the wireless sensor network.

[0052] Optionally, based on the current energy consumption and data collection frequency requirements of each wireless sensor, an optimization algorithm is used in combination with target data to evaluate the optimal working time period required for each wireless sensor, thereby ensuring the efficiency of the wireless sensor network operation and the maximum energy utilization of the wireless sensor nodes.

[0053] Optionally, the target data includes the total working time of the wireless sensor network.

[0054] Optionally, the calculation formula for the total working time of the wireless sensor network is as follows: ; in, represents the total working time of the wireless sensor network, For the The shortest working time of a wireless sensor node under the current energy consumption and data quality requirements, is the total number of nodes in the wireless sensor network.

[0055] Optionally, the working time period of each wireless sensor is dynamically adjusted according to the current energy consumption and data quality requirements of each wireless sensor, so as to optimize energy efficiency and meet data collection requirements, and generate working time period information.

[0056] Optionally, based on the working period information, analyze the impact of differentiated sleep cycles on the performance of the wireless sensor network, evaluate the energy savings of multiple nodes under the target sleep cycle and the impact on data real-time performance, and optimize the sleep strategy of the wireless sensor network by comparing differentiated sleep durations and time periods.

[0057] Optionally, the optimal sleep strategy is determined by calculating the total scores of different sleep strategies.

[0058] Optionally, the total score of the sleep strategy is calculated as follows: ; in, is the total score of the dormancy strategy, For Node Energy savings during the target sleep period, To understand the impact of hibernation on data real-time performance, The weight coefficient representing the energy saving is used to evaluate the importance of energy saving of the node during the target sleep cycle. The weight coefficient representing the impact of data real-time performance is used to evaluate the importance of the negative impact of dormancy on data real-time performance. For the number of nodes considered, the sleep strategy is optimized by evaluating the trade-off between energy savings and data latency, and sleep parameter evaluation data is generated.

[0059] Optionally, the adjusted average sleep parameter of the wireless sensor network is calculated using the following formula: ; in, represents the adjusted average sleep parameter, For Node The energy consumption level, For Node Responsiveness to data quality requirements, The weight coefficient representing the energy consumption level is used to calculate the influence of the energy consumption level of each node in the adjusted sleep parameters. The weight coefficient representing the responsiveness of data quality requirements is used to calculate the influence of the data quality requirements of each node on the sleep parameters in the adjusted sleep parameters. is the total number of nodes, and this formula calculates the optimal sleep parameter for each node by balancing energy consumption and data quality requirements.

[0060] In some embodiments, in step 140, adjusting the data transmission parameters of the plurality of wireless sensors based on the sleep parameters includes: Step 144: Based on the sleep parameters, control the target wireless sensor to sleep, evaluate the transmission performance of the wireless sensor network in real time, and generate network load analysis data; Step 145: Calculate optimal data transmission parameters of multiple wireless sensors based on the network load analysis data and data quality requirements, where the optimal data transmission parameters include optimal data packet capacity and optimal data transmission frequency; Step 146: Based on the optimal data transmission parameters, adjust the data transmission frequency and data packet capacity of the multiple wireless sensors, monitor the communication efficiency and response speed of the wireless sensor network in real time, and obtain the transmission parameter adjustment result.

[0061] Optionally, based on the sleep parameters, considering the impact of wireless sensor node sleep on the transmission performance and network load capacity of the wireless sensor network, the activity status of each node and the total load of the wireless sensor network are evaluated in real time, and a dynamic evaluation method is used to calculate the transmission performance level of the wireless sensor network.

[0062] Optionally, the calculation formula of the transmission performance level of the wireless sensor network is as follows: ; in, Indicates the transmission performance level of the wireless sensor network. For the The activity probability of a node represents the impact of the reduced transmission activity of the node due to sleep. is the total number of nodes in the network, is the base of natural logarithms.

[0063] Optionally, by calculating the product of the activity probabilities of all nodes, the probability that the network can continuously transmit data is obtained, and network load analysis data is generated.

[0064] Optionally, based on the network load analysis data, combined with the real-time and accuracy requirements of the data, the data packet capacity and data transmission frequency of multiple nodes are calculated, and the data packet capacity requirements and transmission frequency requirements of each node are optimized to ensure that the network does not operate overloaded.

[0065] Optionally, taking maximizing the data transmission frequency of the wireless sensor network as the goal, an objective function is constructed, the objective function is solved, and the optimal data transmission parameters of each wireless sensor are calculated.

[0066] Optionally, the objective function is calculated as follows: ; in, is the data transmission frequency of the wireless sensor network, For the The data packet capacity of wireless sensor nodes, For the The transmission time of each wireless sensor node is is the total number of nodes.

[0067] Optionally, the calculation formula of the communication efficiency of the wireless sensor network is as follows: ; in, represents the communication efficiency of the wireless sensor network, For the The transmission frequency of each wireless sensor node is adjusted. For the The transmission frequency of the wireless sensor nodes before adjustment is is the total number of nodes.

[0068] In some embodiments, monitoring the adjusted energy consumption levels of multiple wireless sensors, identifying wireless sensors with abnormal energy consumption, and generating abnormal node detection records in step 150 include: Collect the adjusted energy consumption data of multiple wireless sensors, evaluate the energy consumption change patterns and energy consumption fluctuation ranges of multiple wireless sensors, and generate real-time energy consumption monitoring data; Based on real-time energy consumption monitoring data, wireless sensors with abnormal energy consumption are identified and abnormal node information is generated; Generate warning information based on abnormal node information, send warning information to the user end, and obtain abnormal node detection records.

[0069] Optionally, based on the transmission parameter adjustment result, the energy consumption of the multiple wireless sensors after adjustment is monitored in real time, and time series data of the energy consumption of the multiple wireless sensors after adjustment is collected.

[0070] Optionally, the predicted energy consumption of each wireless sensor is calculated using a pre-built prediction model according to the adjusted energy consumption time series data of each wireless sensor.

[0071] Optionally, the calculation formula of the energy consumption prediction value of each wireless sensor is as follows: ; in, is the predicted energy consumption of each wireless sensor at time t, is the predicted energy consumption of each wireless sensor at time t-1, is the energy consumption of each wireless sensor at time t, is the smoothing coefficient.

[0072] It should be noted that the prediction model is determined based on the historical energy consumption time series data of each wireless sensor. By adjusting the smoothing coefficient, the sensitivity of the prediction model to the historical energy consumption data and the new energy consumption data can be controlled, the real-time changes in energy consumption can be captured and reflected, decision support can be provided for network management, and real-time energy consumption monitoring data can be generated.

[0073] Optionally, based on real-time energy consumption monitoring data, statistical methods are used to identify wireless sensor nodes with abnormal energy consumption, calculate the mean and standard deviation of the energy consumption data of all nodes, analyze the normal fluctuation range of energy consumption, standardize the energy consumption of each node, obtain the standardized energy consumption value of each node, and identify abnormal values ​​that deviate from the normal range.

[0074] Optionally, the calculation formula of the normalized energy consumption value is as follows: ; in, is the standardized energy consumption value of the wireless sensor, is the actual energy consumption value of the wireless sensor, is the average energy consumption of wireless sensors, is the standard deviation of wireless sensor energy consumption.

[0075] Optionally, based on the real-time energy consumption monitoring data, wireless sensors with abnormal energy consumption are identified, the causes of the failures, such as hardware failures or configuration errors, are analyzed, and abnormal node information is generated.

[0076] Optionally, the warning information includes at least the location, operating power consumption and abnormal index of the abnormal node.

[0077] Optionally, a quantitative model is used to assess the severity of the anomaly and determine whether to trigger a notification.

[0078] Optionally, based on the standardized energy consumption value, the probability of sending an early warning message is calculated, and the calculation formula is as follows: ; in, is the probability of sending warning information, is the sensitivity parameter, is the standardized energy consumption value, is the base of natural logarithms.

[0079] It is understandable that by calculating the probability of sending warning information, it can help managers understand the network status in real time, respond quickly to potential problems, and maintain the normal operation of the network.

[0080] Figure 3 A schematic diagram of a process for adjusting the connection paths of multiple wireless sensors based on abnormal node detection records provided by an embodiment of the present invention. Figure 3 As shown, in some embodiments, in step 150, based on the abnormal node detection record, the connection paths of multiple wireless sensors are adjusted to obtain the control parameters of the wireless sensor network, including: Step 151: Based on the abnormal node detection record, monitor the connection data and load information of multiple wireless sensors in real time, evaluate the connection efficiency and overall network load between the multiple wireless sensors, and generate real-time network load information; Step 152: Based on the real-time network load information and the working states of the multiple wireless sensors, as well as the physical distances between the multiple wireless sensors, the Dijkstra algorithm is used to calculate the optimal paths of the multiple wireless sensors, and the connection paths of the multiple wireless sensors are adjusted to generate a connection path adjustment record. Step 153: monitor the transmission performance data of the wireless sensor network in real time, adjust the record and transmission performance data according to the connection path, identify abnormal wireless sensors with connection bottlenecks, and adjust the connection protocol type of the abnormal wireless sensors.

[0081] Optionally, the average load efficiency of the wireless sensor network is calculated using the following formula: ; in, represents the average load efficiency of the wireless sensor network, For the The load percentage of wireless sensor nodes, For the The connection delay of wireless sensor nodes is and is the weight coefficient used to balance the impact of load and delay. is the total number of nodes.

[0082] Optionally, the maximum bottleneck degree of the target wireless sensor is calculated, and whether the target wireless sensor is abnormal is determined according to the maximum bottleneck degree of the target wireless sensor.

[0083] Optionally, the maximum bottleneck degree is calculated as follows: ; in, represents the maximum bottleneck degree identified, For wireless sensor nodes The response time, For wireless sensor nodes service needs.

[0084] In some embodiments, the calculation formula of the optimal path of multiple wireless sensors is as follows: ; in, Represents the distance from the starting wireless sensor node to the target wireless sensor node The shortest path length is used to represent the distance from the source to the node in the network. The optimal path cost is Represents the distance from the starting wireless sensor node to the intermediate wireless sensor node The shortest path length is used to represent the distance from the source to the node in the network. The optimal path cost is For Node To Node The basic weight used to measure the node With Node The direct connection cost, For communication quality The corresponding adjustment coefficient is used to adjust the communication quality The impact on path selection, For Node and nodes The quality of communication between For the current load The corresponding adjustment coefficient is used to adjust the current load The impact on path selection, For Node The current load, For the current load The corresponding adjustment coefficient is used to adjust the current load The impact on path selection, For Node The current load, Logical distance The corresponding adjustment factor is For Node and nodes The logical distance between nodes, used to represent and nodes The network topology or geographical location relationship between them.

[0085] Optionally, obtain the node through the network monitoring system Current load and nodes Current load , measuring the communication quality between nodes , logical distance between nodes , using statistical methods to determine the adjustment factor , , , The value of , substitute the parameter into the formula, combined with the basic weight , calculate the shortest path length from each node to other nodes, optimize the data transmission route in the network, and ensure that the data transmission path is efficient and stable when the network load is high or the communication between nodes is complex.

[0086] Figure 4 A flowchart of a training process of an energy consumption prediction model provided in an embodiment of the present invention is provided. In some embodiments, the training process of the energy consumption prediction model includes: Step 410: Acquire historical operation log data of multiple sample wireless sensors of the sample wireless sensor network, process the historical operation log data, obtain multiple historical working states and multiple historical state conversion times of the multiple sample wireless sensors, and historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Step 420: Determine historical energy consumption analysis results of the plurality of sample wireless sensors based on the historical energy consumption data of the plurality of sample wireless sensors under the plurality of historical working states; Step 430: input multiple historical working states and multiple historical state conversion times of multiple sample wireless sensors into the initial energy consumption prediction model to obtain energy consumption prediction data of multiple sample wireless sensors output by the initial energy consumption prediction model; Step 440: Calculate the loss function value based on the energy consumption prediction data of the multiple sample wireless sensors and the historical energy consumption analysis results of the multiple sample wireless sensors; Step 450: Based on the loss function value, iteratively optimize the parameters of the initial energy consumption prediction model to obtain the energy consumption prediction model.

[0087] It can be understood that by inputting multiple historical working states and multiple historical state conversion times of multiple sample wireless sensors into the initial energy consumption prediction model, the energy consumption prediction data of multiple sample wireless sensors output by the initial energy consumption prediction model are obtained, and based on the energy consumption prediction data of the multiple sample wireless sensors and the historical energy consumption analysis results of the multiple sample wireless sensors, the loss function value is calculated, and the parameters of the initial energy consumption prediction model are iteratively optimized to obtain the energy consumption prediction model without the need for labeling, thereby improving the efficiency of energy consumption prediction model training and the accuracy of energy consumption prediction, and improving the robustness of the energy consumption prediction model.

[0088] In some embodiments, based on the historical energy consumption data of the multiple sample wireless sensors under multiple historical working states, determining the historical energy consumption analysis results of the multiple sample wireless sensors includes: Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical average energy consumption of each sample wireless sensor is calculated; Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical energy consumption level and historical energy consumption fluctuation range of each sample wireless sensor under various historical working conditions are calculated to obtain the historical energy consumption comparison information of each sample wireless sensor; Based on the historical energy consumption comparison information of each sample wireless sensor, the impact of different historical working states and different historical state transition conditions on the historical energy consumption of each sample wireless sensor is evaluated. Combined with the historical average energy consumption of each sample wireless sensor, the historical energy consumption analysis results of each sample wireless sensor are obtained.

[0089] Optionally, the historical average energy consumption of the wireless sensor network under different working conditions is calculated using the following formula: ; in, For wireless sensor networks in working state The average energy consumption under For working status Next The energy consumption of a wireless sensor, For working state The number of wireless sensors under

[0090] The energy control system of the wireless sensor network provided by the embodiment of the present invention is described below. The energy control system of the wireless sensor network described below and the energy control method of the wireless sensor network described above can be referred to each other.

[0091] Figure 5 A schematic diagram of the structure of an energy control system of a wireless sensor network provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, the energy control system 500 of the wireless sensor network includes: An acquisition unit 510 is used to acquire the working status and state transition time of multiple wireless sensors of the wireless sensor network; The prediction unit 520 is used to input the working states and state transition times of the multiple wireless sensors into a pre-built energy consumption prediction model to obtain energy consumption prediction results of the multiple wireless sensors output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on multiple historical working states, multiple historical state transition times and historical energy consumption analysis results of multiple sample wireless sensors; An evaluation unit 530 is used to determine data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluate the data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result; The adjustment unit 540 is used to adjust the sleep cycles of the multiple wireless sensors based on the work demand evaluation result, obtain sleep parameters of the multiple wireless sensors, and adjust the data transmission parameters of the multiple wireless sensors based on the sleep parameters; The monitoring unit 550 is used to monitor the adjusted energy consumption levels of the multiple wireless sensors, identify wireless sensors with abnormal energy consumption, generate abnormal node detection records, and adjust the connection paths of the multiple wireless sensors based on the abnormal node detection records.

[0092] Optionally, based on the abnormal node detection record, adjusting the connection paths of the plurality of wireless sensors includes: Based on abnormal node detection records, the connection data and load information of multiple wireless sensors are monitored in real time, the connection efficiency and overall network load between multiple wireless sensors are evaluated, and real-time network load information is generated; Based on the real-time network load information, the working status of multiple wireless sensors, and the physical distance between multiple wireless sensors, the Dijkstra algorithm is used to calculate the optimal paths of multiple wireless sensors, adjust the connection paths of multiple wireless sensors, and generate connection path adjustment records; Monitor the transmission performance data of the wireless sensor network in real time, adjust the record and transmission performance data according to the connection path, identify abnormal wireless sensors with connection bottlenecks, and adjust the connection protocol type of abnormal wireless sensors.

[0093] Optionally, the calculation formula for the optimal path of multiple wireless sensors is as follows: ; in, Represents the distance from the starting wireless sensor node to the target wireless sensor node The shortest path length, Represents the distance from the starting wireless sensor node to the intermediate wireless sensor node The shortest path length, For Node To Node The basic weight of For communication quality The corresponding adjustment factor is For Node and nodes The quality of communication between For the current load The corresponding adjustment factor is For Node The current load, For the current load The corresponding adjustment factor is For Node The current load, Logical distance The corresponding adjustment factor is For Node and nodes The logical distance between them.

[0094] Optionally, monitoring the adjusted energy consumption levels of the plurality of wireless sensors, identifying wireless sensors with abnormal energy consumption, and generating abnormal node detection records include: Collect the adjusted energy consumption data of multiple wireless sensors, evaluate the energy consumption change patterns and energy consumption fluctuation ranges of multiple wireless sensors, and generate real-time energy consumption monitoring data; Based on real-time energy consumption monitoring data, wireless sensors with abnormal energy consumption are identified and abnormal node information is generated; Generate warning information based on abnormal node information, send warning information to the user end, and obtain abnormal node detection records.

[0095] Optionally, based on the work demand evaluation result, the sleep cycles of the multiple wireless sensors are adjusted to obtain sleep parameters of the multiple wireless sensors, including: Based on the work demand assessment results, determine the data collection frequency requirement of each wireless sensor, and based on the current energy consumption and data collection frequency requirement of each wireless sensor, determine the optimal working time period required for each wireless sensor, and generate working time period information; Based on the working period information, the impact of various sleep cycles on the performance of wireless sensor networks is evaluated, and sleep parameter evaluation data is generated; Based on the sleep parameter evaluation data, considering the energy consumption and data quality requirements of multiple wireless sensors, the sleep parameters of the wireless sensor network are calculated and adjusted to determine the optimal sleep parameters of the wireless sensor network.

[0096] Optionally, based on the sleep parameter, adjusting the data transmission parameters of the plurality of wireless sensors includes: Based on the sleep parameters, the target wireless sensor is controlled to sleep, the transmission performance of the wireless sensor network is evaluated in real time, and network load analysis data is generated; Based on the network load analysis data and data quality requirements, the optimal data transmission parameters of multiple wireless sensors are calculated, and the optimal data transmission parameters include the optimal data packet capacity and the optimal data transmission frequency; Based on the optimal data transmission parameters, the data transmission frequency and data packet capacity of multiple wireless sensors are adjusted, the communication efficiency and response speed of the wireless sensor network are monitored in real time, and the transmission parameter adjustment results are obtained.

[0097] Optionally, according to the energy consumption prediction results of the multiple wireless sensors, data quality requirements of the multiple wireless sensors are determined, and according to the data quality requirements, data collection requirements of the multiple wireless sensors are evaluated to obtain work demand evaluation results, including: Obtain geographic location and network performance indicators of multiple wireless sensors; According to the energy consumption prediction results of the multiple wireless sensors, as well as the geographical locations and network performance indicators of the multiple wireless sensors, the data quality requirements of the multiple wireless sensors are determined, and the data quality requirements include real-time requirements and accuracy requirements; Estimate the power consumption levels of multiple wireless sensors at different data collection frequencies and generate power consumption level analysis results; Based on the power consumption level analysis results and data quality requirements, the optimal data collection frequency parameters of multiple wireless sensors are calculated, and the work demand evaluation results of multiple wireless sensors are obtained.

[0098] Optionally, the training process of the energy consumption prediction model includes: Acquire historical operation log data of multiple sample wireless sensors of the sample wireless sensor network, process the historical operation log data, obtain multiple historical working states and multiple historical state conversion times of the multiple sample wireless sensors, and historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Determine historical energy consumption analysis results of the plurality of sample wireless sensors based on historical energy consumption data of the plurality of sample wireless sensors under a plurality of historical working states; Inputting multiple historical working states and multiple historical state conversion times of multiple sample wireless sensors into an initial energy consumption prediction model to obtain energy consumption prediction data of multiple sample wireless sensors output by the initial energy consumption prediction model; Calculate the loss function value based on the energy consumption prediction data of the multiple sample wireless sensors and the historical energy consumption analysis results of the multiple sample wireless sensors; Based on the loss function value, the parameters of the initial energy consumption prediction model are iteratively optimized to obtain the energy consumption prediction model.

[0099] Optionally, based on the historical energy consumption data of the multiple sample wireless sensors under multiple historical working states, determining the historical energy consumption analysis results of the multiple sample wireless sensors includes: Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical average energy consumption of each sample wireless sensor is calculated; Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical energy consumption level and historical energy consumption fluctuation range of each sample wireless sensor under various historical working conditions are calculated to obtain the historical energy consumption comparison information of each sample wireless sensor; Based on the historical energy consumption comparison information of each sample wireless sensor, the impact of different historical working states and different historical state transition conditions on the historical energy consumption of each sample wireless sensor is evaluated. Combined with the historical average energy consumption of each sample wireless sensor, the historical energy consumption analysis results of each sample wireless sensor are obtained.

[0100] It should be noted here that the energy control system of the wireless sensor network provided in the embodiment of the present invention can implement all the method steps implemented in the above-mentioned energy control method embodiment of the wireless sensor network, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0101] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0102] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for energy control of a wireless sensor network, characterized in that: include: Obtain the working status and state transition time of multiple wireless sensors in a wireless sensor network; Inputting the working states and state transition times of the multiple wireless sensors into a pre-built energy consumption prediction model to obtain energy consumption prediction results of the multiple wireless sensors output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on multiple historical working states, multiple historical state transition times and historical energy consumption analysis results of multiple sample wireless sensors; Determine data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluate data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result; Based on the work demand evaluation result, the sleep cycles of the multiple wireless sensors are adjusted to obtain sleep parameters of the multiple wireless sensors, and based on the sleep parameters, the data transmission parameters of the multiple wireless sensors are adjusted; The adjusted energy consumption levels of the multiple wireless sensors are monitored, wireless sensors with abnormal energy consumption are identified, abnormal node detection records are generated, and connection paths of the multiple wireless sensors are adjusted based on the abnormal node detection records.

2. The energy control method of a wireless sensor network according to claim 1, characterized in that: The adjusting the connection paths of the plurality of wireless sensors based on the abnormal node detection record includes: Based on the abnormal node detection record, the connection data and load information of the multiple wireless sensors are monitored in real time, the connection efficiency and the overall network load between the multiple wireless sensors are evaluated, and the real-time network load information is generated; Based on the real-time network load information and the working states of the multiple wireless sensors, and the physical distances between the multiple wireless sensors, the Dijkstra algorithm is used to calculate the optimal paths of the multiple wireless sensors, adjust the connection paths of the multiple wireless sensors, and generate connection path adjustment records; The transmission performance data of the wireless sensor network is monitored in real time, and abnormal wireless sensors with connection bottlenecks are identified according to the connection path adjustment records and the transmission performance data, and the connection protocol type of the abnormal wireless sensors is adjusted.

3. The energy control method of a wireless sensor network according to claim 2, characterized in that: The calculation formula of the optimal path of the multiple wireless sensors is as follows: ； in, Represents the distance from the starting wireless sensor node to the target wireless sensor node The shortest path length, Represents the distance from the starting wireless sensor node to the intermediate wireless sensor node The shortest path length, For Node To Node The basic weight of For communication quality The corresponding adjustment factor is For Node and nodes The quality of communication between For the current load The corresponding adjustment factor is For Node The current load, For the current load The corresponding adjustment factor is For Node The current load, Logical distance The corresponding adjustment factor is For Node and nodes The logical distance between them.

4. The energy control method of a wireless sensor network according to claim 1, characterized in that: The monitoring of the adjusted energy consumption levels of the plurality of wireless sensors, identifying wireless sensors with abnormal energy consumption, and generating abnormal node detection records includes: Collecting the adjusted energy consumption data of the multiple wireless sensors, evaluating the energy consumption change patterns and energy consumption fluctuation ranges of the multiple wireless sensors, and generating real-time energy consumption monitoring data; Based on the real-time energy consumption monitoring data, identifying wireless sensors with abnormal energy consumption and generating abnormal node information; Generate warning information based on the abnormal node information, send the warning information to the user end, and obtain the abnormal node detection record.

5. The energy control method of a wireless sensor network according to claim 1, characterized in that: The step of adjusting the sleep cycles of the plurality of wireless sensors based on the work demand evaluation result to obtain sleep parameters of the plurality of wireless sensors includes: Based on the work demand evaluation result, determine the data collection frequency requirement of each wireless sensor, based on the current energy consumption and data collection frequency requirement of each wireless sensor, determine the optimal working time period required by each wireless sensor, and generate working time period information; Based on the working period information, evaluating the impact of various sleep cycles on the performance of the wireless sensor network, and generating sleep parameter evaluation data; Based on the sleep parameter evaluation data, the sleep parameters of the wireless sensor network are calculated and adjusted in consideration of the energy consumption and data quality requirements of the multiple wireless sensors, so as to determine the optimal sleep parameters of the wireless sensor network.

6. The energy control method of a wireless sensor network according to claim 1, characterized in that: The adjusting, based on the sleep parameters, the data transmission parameters of the plurality of wireless sensors includes: Based on the sleep parameters, control the target wireless sensor to sleep, evaluate the transmission performance of the wireless sensor network in real time, and generate network load analysis data; Based on the network load analysis data and the data quality requirements, calculating optimal data transmission parameters of the plurality of wireless sensors, the optimal data transmission parameters comprising an optimal data packet capacity and an optimal data transmission frequency; Based on the optimal data transmission parameters, the data transmission frequency and data packet capacity of the multiple wireless sensors are adjusted, the communication efficiency and response speed of the wireless sensor network are monitored in real time, and the transmission parameter adjustment result is obtained.

7. The energy control method of a wireless sensor network according to claim 1, characterized in that: The step of determining data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluating data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain work requirement evaluation results includes: Acquire geographic locations and network performance indicators of the multiple wireless sensors; Determine data quality requirements of the multiple wireless sensors according to energy consumption prediction results of the multiple wireless sensors, as well as geographic locations and network performance indicators of the multiple wireless sensors, wherein the data quality requirements include real-time requirements and accuracy requirements; estimating power consumption levels of the plurality of wireless sensors at differentiated data collection frequencies, and generating power consumption level analysis results; Based on the power consumption level analysis results and data quality requirements, optimal data collection frequency parameters of the multiple wireless sensors are calculated to obtain work requirement evaluation results of the multiple wireless sensors.

8. The energy control method of a wireless sensor network according to claim 1, characterized in that: The training process of the energy consumption prediction model includes: Acquire historical operation log data of multiple sample wireless sensors of the sample wireless sensor network, process the historical operation log data, obtain multiple historical working states and multiple historical state conversion times of the multiple sample wireless sensors, and historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Determine historical energy consumption analysis results of the multiple sample wireless sensors based on historical energy consumption data of the multiple sample wireless sensors under multiple historical working states; Inputting a plurality of historical working states and a plurality of historical state conversion times of the plurality of sample wireless sensors into an initial energy consumption prediction model to obtain energy consumption prediction data of the plurality of sample wireless sensors output by the initial energy consumption prediction model; Calculating a loss function value based on the energy consumption prediction data of the plurality of sample wireless sensors and the historical energy consumption analysis results of the plurality of sample wireless sensors; Based on the loss function value, the parameters of the initial energy consumption prediction model are iteratively optimized to obtain the energy consumption prediction model.

9. The energy control method of a wireless sensor network according to claim 8, characterized in that: The determining of historical energy consumption analysis results of the plurality of sample wireless sensors based on the historical energy consumption data of the plurality of sample wireless sensors under a plurality of historical working states comprises: Calculate the historical average energy consumption of each sample wireless sensor based on the historical energy consumption data of each sample wireless sensor under multiple historical working states; Based on the historical energy consumption data of each sample wireless sensor under various historical working conditions, the historical energy consumption level and the historical energy consumption fluctuation range of each sample wireless sensor under various historical working conditions are calculated to obtain the historical energy consumption comparison information of each sample wireless sensor; Based on the historical energy consumption comparison information of each sample wireless sensor, the impact of different historical working states and different historical state conversion conditions on the historical energy consumption of each sample wireless sensor is evaluated, and combined with the historical average energy consumption of each sample wireless sensor, the historical energy consumption analysis result of each sample wireless sensor is obtained.

10. An energy control system for a wireless sensor network, characterized in that: include: An acquisition unit, used for acquiring the working status and state transition time of multiple wireless sensors of the wireless sensor network; A prediction unit, configured to input the working states and state transition times of the plurality of wireless sensors into a pre-built energy consumption prediction model, and obtain energy consumption prediction results of the plurality of wireless sensors output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on a plurality of historical working states, a plurality of historical state transition times, and historical energy consumption analysis results of a plurality of sample wireless sensors; An evaluation unit, configured to determine data quality requirements of the multiple wireless sensors according to the energy consumption prediction results of the multiple wireless sensors, and evaluate data collection requirements of the multiple wireless sensors according to the data quality requirements to obtain a work requirement evaluation result; An adjusting unit, configured to adjust the sleep cycles of the plurality of wireless sensors based on the work demand evaluation result, obtain sleep parameters of the plurality of wireless sensors, and adjust the data transmission parameters of the plurality of wireless sensors based on the sleep parameters; The monitoring unit is used to monitor the adjusted energy consumption levels of the multiple wireless sensors, identify wireless sensors with abnormal energy consumption, generate abnormal node detection records, and adjust the connection paths of the multiple wireless sensors based on the abnormal node detection records.

Images