A substation environment anomaly early warning method based on an internet of things

Abstract

The present application relates to substation environment early warning technical field, specifically disclose a kind of substation environment abnormal early warning method based on internet of things, the present application is first based on boundary element model, the data obtained by a monitoring point located on boundary can be obtained by the way of boundary element solution calculation to obtain the sound pressure data of remaining nodes on boundary, and traditional sound monitoring, in order to be able to find various fault conditions in time, multiple monitoring devices need to be set to a sound source device, that is, the present application greatly reduces the use of monitoring device, in addition, by judging whether the free field region is coincident in the sound diffusion process of each sound source device, in the case of coincidence, the sound source equipment of two substations is monitored simultaneously by a monitoring device, further reduce the number of monitoring devices, so that only a few monitoring devices can realize the real-time monitoring of noise distribution of entire substation, the cost of operation and maintenance is low, and it is convenient to install in the operation station.

Description

Technical Field

[0001] This invention relates to the field of substation environmental early warning technology, and more specifically, to a method for early warning of substation environmental anomalies based on the Internet of Things. Background Technology

[0002] With increasing power demand, grid capacity, and rising public awareness of environmental protection, substation noise is drawing growing attention. Especially during peak summer power supply periods, urban substations receive numerous complaints due to noise pollution, leading to significant resource expenditure on investigation, responses, and coordination, potentially causing public opinion incidents and impacting "power supply service satisfaction." For substations with frequent and severe noise complaints, the financial investment in noise reduction upgrades is substantial. Furthermore, there are currently no specialized calculation tools for evaluating noise reduction plans and their effectiveness. Commonly used noise prediction software in power design departments employs algorithms unsuitable for the multi-phase sound sources and complex outdoor propagation scenarios of substations, resulting in significant errors in the final results. Therefore, upgrade recommendations based on these commonly used algorithms tend to be conservative and have large margins, often leading to budget overruns. The large error also prevents these commonly used algorithms from being used for accurate online monitoring.

[0003] The significance of online monitoring lies in the fact that the main risks of excessive noise in old urban substations stem from equipment aging, abnormal fan noise, and fluctuations in operating conditions (such as DC bias). If the peak decibel level of noise can be predicted during the substation's commissioning process, potential noise hazards can be identified in a timely manner, allowing for preventative measures to be taken in advance during off-peak electricity seasons through power outages for inspection and renovation.

[0004] In view of this, the present invention proposes an Internet of Things-based method for early warning of environmental anomalies in substations, which is used for online monitoring of noise in scenarios with multiple coherent sound sources and complex outdoor propagation in substations. Summary of the Invention

[0005] The purpose of this invention is to provide an early warning method for substation environmental anomalies based on the Internet of Things, solving the following technical problems:

[0006] How to conduct online noise monitoring in scenarios involving multiple coherent sound sources in substations and complex outdoor propagation?

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] A method for early warning of environmental anomalies in substations based on the Internet of Things includes the following steps:

[0009] S100: Based on the substation environment, perform environmental modeling to obtain a spatial model, and at the same time, use the Internet of Things to obtain the power load of the substation's sound source equipment.

[0010] S200. Establish a holographic virtual sound source model of the substation sound source equipment, as well as a composite sound channel model that implicitly contains information such as sound barriers and buildings;

[0011] S300 combines a holographic virtual sound source model and a composite sound channel model with a spatial model, and constructs an acoustic boundary element model for each sound source device.

[0012] S400. Determine several monitoring points based on the boundary element model of the sound source equipment;

[0013] S500: Noise monitoring is performed on substation equipment at monitoring points using monitoring equipment. At the same time, boundary element method is solved based on the noise data received by the monitoring equipment to obtain sound pressure data of multiple nodes on the boundary of the boundary element model.

[0014] S600 determines whether to issue an early warning for the substation environment based on the sound pressure data and power load of multiple nodes of the current sound source equipment.

[0015] As a further technical solution of the present invention: the implementation process of step S400 includes the following steps:

[0016] S410. Obtain the free field region of the sound source device during sound propagation based on the parameters of the sound source device provided by the manufacturer.

[0017] S420. In the spatial model, locate each free field region based on the location information of the sound source device, and determine whether there are intersecting free field regions.

[0018] S430. Set up a monitoring point in the overlapping part of the intersecting free field regions, and set up a monitoring point in each free field that does not intersect with other free field regions.

[0019] The above technical solution provides a sound monitoring solution using a small number of monitoring devices. Specifically, this invention is based on the boundary element model. Data obtained from a monitoring point located on the boundary can be used to calculate the sound pressure data of the remaining nodes on the boundary through boundary element solving. In contrast, traditional sound monitoring requires multiple monitoring devices for a single sound source device in order to detect various fault states in a timely manner. In other words, the above process greatly reduces the number of monitoring devices used. Furthermore, by determining whether the free field regions of each sound source device overlap during sound diffusion, if they overlap, a single monitoring device can simultaneously monitor the sound source devices of two substations, further reducing the number of monitoring devices. Thus, only a few monitoring devices are needed to achieve real-time monitoring of the noise distribution of the entire substation, resulting in low operation and maintenance costs and easy installation at commissioning stations.

[0020] As a further technical solution of the present invention: In step S500, the process of solving the boundary element problem includes:

[0021] S510. Obtain the boundary element model of the current sound source device and the location of the monitoring point corresponding to the current sound source device;

[0022] S520. Establish boundary element equations based on sound pressure data, and the boundary of the boundary element equations passes through the detection point positions.

[0023] S530. Discretize the boundary of the boundary element equation to obtain multiple nodes, and solve the boundary element equation based on the noise data of the monitoring points to obtain multiple node solutions.

[0024] As a further technical solution of the present invention, the steps for providing early warning of the substation environment include:

[0025] S610. Based on the noise exceeding risk items of the current sound source equipment, set several reference points from multiple nodes so that each noise exceeding risk item has at least one reference point.

[0026] S620. Obtain the electrical load of the sound source equipment from the electrical load, and establish several coordinate systems with the electrical load as the horizontal axis and the sound pressure data as the vertical axis.

[0027] S630. Record the data set of a noise exceeding risk project during the operation of a sound source device in a coordinate system, and form a two-dimensional data distribution map.

[0028] S640. After completing the recording, a two-dimensional data distribution map is generated for each noise exceeding risk item, and regression analysis is performed to obtain the regression curve corresponding to the noise exceeding risk item.

[0029] S650. Based on the regression curves corresponding to the noise exceeding risk items, evaluate each noise exceeding risk item to obtain the evaluation coefficient, and determine whether to issue an early warning for the substation environment based on the evaluation coefficient.

[0030] As a further technical solution of the present invention: the process of obtaining the evaluation coefficient includes:

[0031] For a total of m projects, a risk interval is set within the horizontal axis range of the regression curve and standard curve corresponding to each project with noise exceeding the standard risk. This risk interval is then divided into n equal evaluation intervals. Finally, the following formula is used:

[0032]

[0033]

[0034] Obtain the evaluation coefficient Ev j Among them, Ev j A is the evaluation coefficient of the j-th item among m noise exceedance risk items.i S is the comparison value of the i-th evaluation interval in order among n evaluation intervals. i and S i+1 Let S be the electrical load of the current sound source device, S be the two endpoint values ​​of the i-th evaluation interval, F1 be the regression curve function, F0 be the standard curve function, and K be the standard curve function. s It is the slope of the evaluation interval on the regression curve that is closest to the right endpoint of the risk interval, and the slope of the evaluation interval on the K0 standard curve that is closest to the right endpoint of the risk interval. r is a preset function.

[0035] As a further technical solution of the present invention: the step of determining whether to issue an early warning for the substation environment based on the evaluation coefficient includes:

[0036] S651. Obtain the power load change curve, and set a safety threshold range for each noise exceedance risk item when the power load continues to rise.

[0037] S652, the evaluation coefficient Ev of the j-th noise exceedance risk item. j and the corresponding safety threshold range [E j E j+1 Compare;

[0038] S653, if Ev j >E j+1 If the j-th noise exceedance risk item of the current sound source equipment is recorded as qualified, then if Ev j <E j If the j-th noise level of the current sound source device exceeds the standard, then the current noise level exceeding the standard risk item is recorded as unqualified. If Ev j ∈[E j E j+1 If the j-th noise level of the current sound source device exceeds the standard, then the critical state is recorded as the j-th noise level exceeding the standard.

[0039] S654. If all m noise risk items of the current sound source equipment are qualified, no warning will be issued; if all m noise risk items of the current sound source equipment are unqualified, a warning will be issued; otherwise, a second judgment will be made to decide whether to issue a warning.

[0040] The above technical solution provides a process for obtaining evaluation coefficients and determining whether to issue an early warning for the substation environment based on these coefficients. Specifically, the evaluation coefficients of this invention are influenced by the comparison value and the ratio between the slope of the evaluation interval closest to the right endpoint of the risk interval on the regression curve and the slope of the evaluation interval closest to the right endpoint of the risk interval on the standard curve. Alternatively, the larger the comparison value, the larger the evaluation coefficient. By analyzing the dangerous range, it is possible to predict the state of the upcoming peak electricity consumption during the continuous rise of electricity load. Long-term sound field monitoring can promptly detect noise source equipment that exceeds the noise standard, thereby allowing sufficient maintenance time to eliminate hidden dangers.

[0041] As a further technical solution of the present invention: the secondary judgment process includes:

[0042] Through the formula:

[0043]

[0044] Obtain the judgment coefficients Mq and σ j is the weighting coefficient of the j-th noise exceeding risk item, and R is a preset piecewise function;

[0045] Determine if Mq is greater than M1. If so, issue an alert; otherwise, do not issue an alert. M1 is a preset threshold value.

[0046] The above technical solution provides a secondary judgment process. Specifically, the secondary judgment of this invention is carried out through weighted voting calculation, and the set piecewise function filters out noise exceeding risk items that meet the voting criteria by limiting the threshold value. This enables the comprehensive evaluation of multiple evaluation coefficients with different criteria, thereby unifying the comparison scale and improving the accuracy of the secondary judgment.

[0047] As a further technical solution of the present invention: the risk range is set within the range where the electricity load is continuously increasing, and the risk range is set before the peak load of the summer peak electricity consumption period.

[0048] The beneficial effects of this invention are:

[0049] (1) This invention incorporates a modular holographic virtual sound source model of sound source equipment into the environmental model of the substation. Combined with a composite sound channel model, it forms an acoustic boundary element model to calculate the noise field distribution inside and outside the substation in real time. This model can predict the peak noise during the peak summer season. In addition, noise monitoring and reproduction can automatically adapt to the impact of new sound source equipment such as operating condition fluctuations and wind turbine commissioning, and predict the peak noise during the peak summer season. At the same time, through long-term sound field monitoring, it can also promptly detect sound source equipment with excessive noise and accurately push technical improvement suggestions and maintenance reminders for abnormal noise source equipment.

[0050] (2) The present invention reduces the use of monitoring equipment. Furthermore, by determining whether the free field regions of each sound source device overlap during the sound diffusion process, and in the case of overlap, a single monitoring device can simultaneously monitor the sound source devices of two substations, further reducing the number of monitoring devices. Thus, only a few monitoring devices are needed to achieve real-time monitoring of the noise distribution of the entire substation. The operating and maintenance costs are low, and it is easy to install at the commissioning station.

[0051] (3) By analyzing the dangerous zone, this invention can predict the state of the upcoming peak electricity consumption during the continuous rise of electricity load, and can promptly detect the noise source equipment that exceeds the standard through long-term sound field monitoring, thereby leaving enough maintenance time to eliminate hidden dangers. Attached Figure Description

[0052] The invention will now be further described with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart of the early warning method of the present invention;

[0054] Figure 2 This is a flowchart illustrating the steps of the present invention for providing early warning of the substation environment. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Please see Figure 1 As shown in one embodiment, a substation environmental anomaly early warning method based on the Internet of Things is provided, including the following steps:

[0057] S100. Based on the substation environment, environmental modeling is performed to obtain a spatial model. At the same time, the electrical load of the substation's sound source equipment is obtained through the Internet of Things. The spatial model is obtained by three-dimensional reconstruction. The specific method is not described in detail.

[0058] The S200 system maps the near-field frequency response and global environmental frequency response matrix of the sound source, establishing a holographic virtual sound source model for the substation's sound source equipment. This model describes point-like sound sources and includes a composite sound channel model that implicitly incorporates information about sound barriers, buildings, etc. The holographic virtual sound source model and composite sound channel model are used to recreate scenes and achieve digital twins. The holographic virtual sound source model typically employs the following steps: 1. Sound Acquisition: Using multiple microphones or other sound acquisition devices, sound emitted from the sound source is acquired from different angles and distances. These sound signals are recorded and stored in a computer. 2. Sound Signal Processing: Using computer algorithms and programs, the acquired sound signals are processed and analyzed. This includes sound source localization, sound propagation path analysis, sound reflection and refraction, etc. 3. Three-Dimensional Sound Reconstruction: Based on the processed and analyzed sound signals, computer graphics technology is used to reconstruct a sound with a three-dimensional spatial sense in the computer. This sound effect allows users to perceive the location, distance, direction, and other characteristics of the sound source, as well as the effects of sound propagation and reflection in space.

[0059] A composite channel model is a model describing complex sound systems, composed of multiple simple channel models. Each channel model can be a simple sound propagation path, such as a room, a pipe, or a bridge. In a composite channel model, each channel has its own physical properties and sound propagation characteristics, such as length, diameter, shape, and material. Each channel has different effects on sound, such as reflection, refraction, and absorption. By combining multiple channels, composite channel models can simulate complex sound environments, such as cities, mountains, and buildings. In this model, sound propagates from one channel to another and is influenced and altered by each channel.

[0060] S300: Based on the combination of holographic virtual sound source model and composite sound channel model with spatial model, an acoustic boundary element model is constructed for each sound source device. The boundary element model is usually set up for sound propagation in solids. The noise in the environment in which this invention is used is low-frequency noise. Low-frequency noise decreases very slowly and the sound wave is relatively long, so it can easily pass through obstacles. Therefore, the boundary element model is established with a point as the center, and there will be no excessive error. The boundary line obtained by controlling the variance approximation can also represent the sound pressure propagation state of the sound.

[0061] S400. Determine several monitoring points based on the boundary element model of the sound source equipment;

[0062] S500: Noise monitoring is performed on substation equipment at monitoring points using monitoring equipment. At the same time, boundary element method is solved based on the noise data received by the monitoring equipment to obtain sound pressure data of multiple nodes on the boundary of the boundary element model.

[0063] S600 determines whether to issue an early warning for the substation environment based on the sound pressure data and power load of multiple nodes of the current sound source equipment.

[0064] This embodiment provides a method for online detection of substation noise environment. Specifically, the invention incorporates a modular holographic virtual sound source model of sound source equipment into the substation environment model, and combines it with a composite sound channel model to form an acoustic boundary element model to calculate the noise field distribution inside and outside the substation in real time. This can predict the peak noise during the "peak summer" period. In addition, noise monitoring and reproduction can automatically adapt to the impact of operating condition fluctuations, wind turbine commissioning, and other newly added sound source equipment, predicting the peak noise during the "peak summer" period. Simultaneously, through long-term sound field monitoring, it can also promptly identify sound source equipment exceeding noise standards and accurately push technical improvement suggestions and maintenance reminders for abnormal noise source equipment.

[0065] It should be noted that the technical upgrade plan affects the composite sound channel model. In other words, the composite sound channel model is modified before the technical upgrade plan is implemented in order to make predictions and provide reasonable suggestions.

[0066] The implementation process of step S400 includes the following steps:

[0067] S410. Obtain the free field region of the sound source device during sound propagation based on the parameters of the sound source device provided by the manufacturer.

[0068] S420. In the spatial model, locate each free field region based on the location information of the sound source device, and determine whether there are intersecting free field regions.

[0069] S430. Set up a monitoring point in the overlapping part of the intersecting free field regions, and set up a monitoring point in each free field that does not intersect with other free field regions.

[0070] This embodiment provides a technical solution for sound monitoring using a small number of monitoring devices. Specifically, the present invention is based on the boundary element model. The sound pressure data of the remaining nodes on the boundary can be calculated by using the data obtained from a monitoring point located on the boundary. In contrast, traditional sound monitoring requires multiple monitoring devices for a single sound source device in order to detect various fault states in a timely manner. In other words, the above process greatly reduces the number of monitoring devices used. In addition, by judging whether the free field regions of each sound source device overlap during the sound diffusion process, if they overlap, a single monitoring device can simultaneously monitor the sound source devices of two substations, further reducing the number of monitoring devices. Thus, only a few monitoring devices are needed to achieve real-time monitoring of the noise distribution of the entire substation, resulting in low operation and maintenance costs and easy installation at the commissioning station.

[0071] In step S500, the process of solving the boundary element problem includes:

[0072] S510. Obtain the boundary element model of the current sound source device and the location of the monitoring point corresponding to the current sound source device;

[0073] S520. Establish boundary element equations based on sound pressure data, and the boundary of the boundary element equations passes through the detection point positions.

[0074] S530. Discretize the boundary of the boundary element equation to obtain multiple nodes, and solve the boundary element equation based on the noise data of the monitoring points to obtain multiple node solutions.

[0075] refer to Figure 2 The steps for issuing early warnings about the substation environment include:

[0076] S610. Based on the noise exceeding risk items of the current sound source equipment, set several reference points from multiple nodes so that each noise exceeding risk item has at least one reference point.

[0077] S620. Obtain the electrical load of the sound source equipment from the electrical load, and establish several coordinate systems with the electrical load as the horizontal axis and the sound pressure data as the vertical axis.

[0078] S630. During the operation of the sound source equipment, a data set of a noise exceeding the standard risk item is collected in a coordinate system to form a two-dimensional data distribution map. The horizontal axis of the data set is the current power load, and the vertical axis is the sound pressure data of the reference point corresponding to the noise exceeding the standard risk item at the current time.

[0079] S640. After completing the recording, a two-dimensional data distribution map is generated for each noise exceeding risk item, and regression analysis is performed to obtain the regression curve corresponding to the noise exceeding risk item.

[0080] S650. Based on the regression curves corresponding to the noise exceeding risk items, evaluate each noise exceeding risk item to obtain the evaluation coefficient, and determine whether to issue an early warning for the substation environment based on the evaluation coefficient.

[0081] The process of obtaining the evaluation coefficient includes:

[0082] For a total of m projects, a risk interval is set within the horizontal axis range of the regression curve and standard curve corresponding to each project with noise exceeding the standard risk. This risk interval is then divided into n equal evaluation intervals. Finally, the following formula is used:

[0083]

[0084]

[0085] Obtain the evaluation coefficient Ev j Among them, Ev jA is the evaluation coefficient of the j-th item among m noise exceedance risk items. i It is the comparison value of the i-th evaluation interval in n evaluation intervals, arranged in ascending order based on the value of the interval endpoints in this embodiment. S i and S i+1 Let S be the two endpoint values ​​of the i-th evaluation interval, S be the current electrical load of the sound source device, F1 be the regression curve function, F0 be the standard curve function, and the standard curve is preset based on normal operating conditions. s is the slope of the evaluation interval on the regression curve that is closest to the right endpoint of the risk interval, and the slope of the evaluation interval on the K0 standard curve that is closest to the right endpoint of the risk interval. r is a preset monotonically increasing transformation function, and in this embodiment, a lookup table function is used.

[0086] The steps for determining whether to issue an early warning for the substation environment based on the evaluation coefficient include:

[0087] S651. Obtain the power load change curve, and set a safety threshold range for each noise exceedance risk item when the power load continues to rise.

[0088] S652, the evaluation coefficient Ev of the j-th noise exceedance risk item. j and the corresponding safety threshold range [E j E j+1 Compare;

[0089] S653, if Ev j >E j+1 If the j-th noise exceedance risk item of the current sound source equipment is recorded as qualified, then if Ev j <E j If the j-th noise level of the current sound source device exceeds the standard, then the current noise level exceeding the standard risk item is recorded as unqualified. If Ev j ∈[E j E j+1 If the j-th noise level of the current sound source device exceeds the standard, then the critical state is recorded as the j-th noise level exceeding the standard.

[0090] S654. If all m noise risk items of the current sound source equipment are qualified, no warning will be issued; if all m noise risk items of the current sound source equipment are unqualified, a warning will be issued; otherwise, a second judgment will be made to decide whether to issue a warning.

[0091] This embodiment provides a process for obtaining evaluation coefficients and determining whether to issue an early warning for the substation environment based on these coefficients. Specifically, the evaluation coefficients of this invention are influenced by the comparison value and the ratio between the slope of the evaluation interval closest to the right endpoint of the risk interval on the regression curve and the slope of the evaluation interval closest to the right endpoint of the risk interval on the standard curve. Alternatively, the larger the comparison value, the larger the evaluation coefficient. By analyzing the dangerous range, it is possible to predict the state of the upcoming peak electricity consumption during the continuous rise of electricity load. Long-term sound field monitoring can promptly detect noise source equipment that exceeds the noise standard, thereby allowing sufficient maintenance time to eliminate hidden dangers.

[0092] The process of secondary judgment includes:

[0093] Through the formula:

[0094]

[0095] Obtain the judgment coefficients Mq and σ j is the weight coefficient of the j-th noise exceeding risk item, which is a preset value. R is a preset piecewise function. In the range [0, 1), R outputs 0. In the range greater than 1, R is a normalization function.

[0096] Determine if Mq is greater than M1. If so, issue an alert; otherwise, do not issue an alert. M1 is a preset threshold value.

[0097] This embodiment provides a secondary judgment process. Specifically, the secondary judgment of the present invention is carried out through weighted voting calculation, and the set piecewise function filters out noise exceeding risk items that meet the voting criteria by limiting the threshold value. This enables the comprehensive evaluation coefficients of multiple different criteria to unify the comparison scale and improve the accuracy of the secondary judgment.

[0098] The risk zone is set within the range where electricity load increases continuously, and the risk zone is set before the peak load of the summer peak electricity consumption period.

[0099] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A substation environment anomaly early warning method based on Internet of Things, characterized in that, Includes the following steps: S100: Based on the substation environment, perform environmental modeling to obtain a spatial model, and at the same time, use the Internet of Things to obtain the power load of the substation's sound source equipment. S200. Establish a holographic virtual sound source model of the substation sound source equipment, as well as a composite sound channel model that implicitly contains information such as sound barriers and buildings; S300 combines a holographic virtual sound source model and a composite sound channel model with a spatial model, and constructs an acoustic boundary element model for each sound source device. S400. Determine several monitoring points based on the boundary element model of the sound source equipment; S500: Noise monitoring is performed on substation equipment at monitoring points using monitoring equipment. At the same time, boundary element method is solved based on the noise data received by the monitoring equipment to obtain sound pressure data of multiple nodes on the boundary of the boundary element model. S600: Based on the sound pressure data and power load of multiple nodes of the current sound source equipment, determine whether to issue an early warning for the substation environment; The process of obtaining the evaluation coefficient includes: In total For each noise over-standard risk item, the risk interval is set in the horizontal coordinate range of the regression curve and the standard curve corresponding to the noise over-standard risk item, the risk interval is equally divided into Then, the formula is: ； ； Obtaining evaluation coefficient ,in, yes The first project with noise exceeding risk The evaluation coefficient of each project yes In the evaluation intervals, the first one in order The comparison value of each evaluation interval. For the first The two endpoint values ​​of an evaluation interval This is the current electrical load of the sound source equipment. It is the regression curve function. It is a standard curve function. It is the slope of the evaluation interval on the regression curve that is closest to the right endpoint of the risk interval. The slope of the evaluation interval on the standard curve that is closest to the right endpoint of the risk interval. It is a preset conversion function; The steps for determining whether to issue an early warning for the substation environment based on the evaluation coefficient include: S651. Obtain the power load change curve, and set a safety threshold range for each noise exceedance risk item when the power load continues to rise. S652, the first Evaluation coefficient of each noise exceedance risk item and the corresponding safety threshold range Perform a comparison; S653, if Then the current sound source device's first Each noise level exceeding the risk level is recorded as qualified. Then the current sound source device's first Each item with a risk of exceeding noise standards is marked as unqualified. Then the current sound source device's first Each project with a risk of exceeding noise standards is recorded as being in a critical state. S654, In the current sound source device No warning will be issued if all noise level risk items are within acceptable limits, given the current noise source equipment. If all noise-risk items are found to be non-compliant, an early warning will be issued; otherwise, a second assessment will be conducted to determine whether to issue an early warning. The process of secondary judgment includes: Through the formula: ； Obtain the judgment coefficient , It is the first The weighting coefficient of each noise exceeding risk item It is a pre-defined piecewise function; judge Is it greater than If so, then a warning needs to be issued; otherwise, no warning is issued. It is a preset critical value; The risk range is set within the range where the electricity load is continuously increasing, and the risk range is set before the peak load of the summer peak electricity consumption period.

2. The method for early warning of substation environmental anomalies based on the Internet of Things according to claim 1, characterized in that, The implementation process of step S400 includes the following steps: S410. Obtain the free field region of the sound source device during sound propagation based on the parameters of the sound source device provided by the manufacturer. S420. In the spatial model, locate each free field region based on the location information of the sound source device, and determine whether there are intersecting free field regions. S430. Set up a monitoring point in the overlapping part of the intersecting free field regions, and set up a monitoring point in each free field that does not intersect with other free field regions.

3. The method for early warning of substation environmental anomalies based on the Internet of Things according to claim 2, characterized in that, In step S500, the process of solving the boundary element problem includes: S510. Obtain the boundary element model of the current sound source device and the location of the monitoring point corresponding to the current sound source device; S520. Establish the boundary element equation based on sound pressure data, and the boundary of the boundary element equation passes through the detection point position. S530. Discretize the boundary of the boundary element equation to obtain multiple nodes, and solve the boundary element equation based on the noise data of the monitoring points to obtain the number of solutions for multiple nodes.

4. The method for early warning of substation environmental anomalies based on the Internet of Things according to claim 3, characterized in that, The steps for issuing early warnings about the substation environment include: S610. Based on the noise exceeding risk items of the current sound source equipment, set several reference points from multiple nodes so that each noise exceeding risk item has at least one reference point. S620. Obtain the electrical load of the sound source equipment from the electrical load, and establish several coordinate systems with the electrical load as the horizontal axis and the sound pressure data as the vertical axis. S630. Record the data set of a noise exceeding risk project during the operation of a sound source device in a coordinate system, and form a two-dimensional data distribution map. S640. After completing the recording, a two-dimensional data distribution map is generated for each noise exceeding risk item, and regression analysis is performed to obtain the regression curve corresponding to the noise exceeding risk item. S650. Based on the regression curves corresponding to the noise exceeding risk items, evaluate each noise exceeding risk item to obtain the evaluation coefficient, and determine whether to issue an early warning for the substation environment based on the evaluation coefficient.

Images