A dynamic risk assessment method and system based on the fusion of multi-source heterogeneous data

Abstract

The present invention provides a dynamic risk assessment method and system for fusing multi-source heterogeneous data, wherein the method comprises: filtering environmental parameter signals to obtain filtered environmental parameter signals; parsing the filtered environmental parameter signals to obtain environmental parameters; constructing an outlier screening matrix according to the environmental parameters; removing outliers using the outlier screening matrix to obtain a multi-source heterogeneous data fusion set; fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value; and issuing a construction risk warning when the construction risk assessment value is greater than an alarm threshold. The present invention can make the environmental parameters collected by the sensor more accurate by parsing the filtered environmental parameter signals to obtain environmental parameters, and then fusing all environmental parameters based on the multi-source heterogeneous data fusion set to obtain a construction risk assessment value, which can comprehensively improve the accuracy of the alarm information by combining multiple factors.

Description

Technical Field

[0001] The present invention belongs to the technical field of construction early warning, and more specifically, relates to a dynamic risk assessment method and system integrating multi-source heterogeneous data. Background Art

[0002] In recent years, with the development of economy and society, the status of infrastructure industry in my country's national economy has been increasing. However, as the centralized implementation place of construction production factors, the working environment of infrastructure construction site is constantly changing, and there are great safety hazards. On the one hand, the site environment conditions of infrastructure construction site are complex and changeable, and there are a large number of open-air operations and high-altitude operations; the construction process involves a variety of operation processes such as construction components, construction machinery, operating equipment, and transportation vehicles. The current technical conditions make it difficult to effectively identify, track and monitor the various unsafe factors in the site environment. On the other hand, the construction workers are densely populated and the personnel mobility is high. The construction process also involves comprehensive cross-operation of multiple types of work and the operation of large-scale machinery and equipment. Mechanical construction and manual operation coexist.

[0003] With the continuous development of the computer industry, detection and early warning of personnel in dangerous areas by placing a single sensor has been applied to most industries with dangerous operations; however, the above technology can only provide early warning for local areas of the construction site, and the early warning accuracy is not high. Summary of the invention

[0004] The purpose of the present invention is to provide a dynamic risk assessment method and system integrating multi-source heterogeneous data, aiming to solve the problem of poor early warning effect at construction sites.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is: a dynamic risk assessment method integrating multi-source heterogeneous data, comprising:

[0006] Step 1: Obtain environmental parameter signals collected by various sensors at the construction site;

[0007] Step 2: Filtering the environmental parameter signal to obtain a filtered environmental parameter signal;

[0008] Step 3: Analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration;

[0009] Step 4: construct an outlier screening matrix based on the environmental parameters;

[0010] Step 5: Use the outlier screening matrix to remove outliers to obtain a multi-source heterogeneous data fusion set;

[0011] Step 6: Fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value;

[0012] Step 7: When the construction risk assessment value is greater than the alarm threshold, a construction risk warning is issued.

[0013] Preferably, the step 2: filtering the environmental parameter signal to obtain a filtered environmental parameter signal comprises:

[0014] Step 2.1: performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients;

[0015] Step 2.2: Calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is:

[0016]

[0017] Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale;

[0018] Step 2.3: construct a self-adjusting threshold according to the noise standard deviation;

[0019] Step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold;

[0020] Step 2.5: Filter the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal.

[0021] Preferably, the step 2.3: constructing a self-adjusting threshold according to the noise standard deviation comprises:

[0022] Using the formula:

[0023] δ i =In(e+2 1-i -1)×σ i

[0024] Construct a self-regulating threshold; where δ i Represents the self-adjustment threshold at the i-th decomposition scale.

[0025] Preferably, the step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold value comprises:

[0026] Using the formula:

[0027]

[0028] Construct a self-regulating filter function; where W δirepresents the wavelet coefficient after filtering, W ij represents the jth wavelet coefficient under the i-th decomposition scale, k represents the adjustment factor, T represents the adaptive coefficient, and its value range is [1, 3].

[0029] Preferably, the step 4: constructing an outlier screening matrix according to the environmental parameters comprises:

[0030] Step 4.1: Construct the trust characteristic equation of each sensor according to the variance of the environmental parameter; wherein the trust characteristic equation is:

[0031]

[0032] In the formula, p i (x / x i ) represents the trust characteristic equation of the i-th sensor, x i represents the environmental parameters collected by the i-th sensor at time t, σ i represents the standard deviation of the environmental parameters collected by sensor i;

[0033] Step 4.2: Construct an outlier screening matrix based on the trust characteristic equation.

[0034] Preferably, the step 4.2: constructing an outlier screening matrix according to the trust characteristic equation comprises:

[0035] Using the formula:

[0036]

[0037] Construct an outlier screening matrix; where, represents the distance measure between sensor i and sensor j.

[0038] Preferably, the step 5: using the outlier screening matrix to remove outliers to obtain a multi-source heterogeneous data fusion set includes:

[0039] Step 5.1: removing the element values ​​in the outlier screening matrix that are greater than a preset threshold value to obtain a credibility matrix;

[0040] Step 5.2: Extract the corresponding environmental parameters in the credibility matrix to obtain a multi-source heterogeneous data fusion set.

[0041] Preferably, the step 6: fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value includes:

[0042] Using the formula:

[0043]

[0044] The data in the multi-source heterogeneous data fusion set is integrated to obtain the construction risk assessment value; among them, represents the construction risk assessment value, W i represents the fusion weight of the i-th sensor, X i Represents the data collected by the i-th sensor in the multi-source heterogeneous data fusion set.

[0045] The present invention also provides a dynamic risk assessment system integrating multi-source heterogeneous data, comprising:

[0046] An environmental parameter signal acquisition module is used to acquire environmental parameter signals collected by various sensors at the construction site;

[0047] A filtering module, used for filtering the environmental parameter signal to obtain a filtered environmental parameter signal;

[0048] An environmental parameter analysis module, used to analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration;

[0049] A screening matrix construction module, used to construct an outlier screening matrix according to the environmental parameters;

[0050] A data fusion set screening module, used for removing outliers by using the outlier screening matrix to obtain a multi-source heterogeneous data fusion set;

[0051] A fusion module, used for fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value;

[0052] The early warning module is used to issue a construction risk early warning when the construction risk assessment value is greater than the alarm threshold.

[0053] Preferably, the filtering module comprises:

[0054] A wavelet decomposition unit, used for performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients;

[0055] A noise standard deviation calculation unit is used to calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is:

[0056]

[0057] Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale;

[0058] A self-adjusting threshold value construction unit, used to construct a self-adjusting threshold value according to the noise standard deviation;

[0059] A self-adjusting filter function construction unit, used to construct a self-adjusting filter function according to the self-adjusting threshold;

[0060] A filtering unit is used to filter the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal.

[0061] The beneficial effect of a dynamic risk assessment method and system for fusing multi-source heterogeneous data provided by the present invention is that: compared with the prior art, the present invention can make the environmental parameters collected by the sensor more accurate by analyzing the filtered environmental parameter signals to obtain environmental parameters, and then fuse all environmental parameters based on the multi-source heterogeneous data fusion set to obtain a construction risk assessment value, which can comprehensively consider multiple factors to improve the accuracy of the alarm information. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only 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.

[0063] Figure 1 A flow chart of a dynamic risk assessment method for fusing multi-source heterogeneous data provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of a dynamic risk assessment system for fusing multi-source heterogeneous data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] The purpose of the present invention is to provide a dynamic risk assessment method and system integrating multi-source heterogeneous data, aiming to solve the problem of poor early warning effect at construction sites.

[0067] See also Figure 1 , a dynamic risk assessment method integrating multi-source heterogeneous data, including:

[0068] Step 1: Obtain environmental parameter signals collected by various sensors at the construction site;

[0069] In practical applications, the present invention firstly needs to arrange a plurality of sensors at the construction site to collect different environmental parameter signals, and then upload them to the host computer by wireless transmission.

[0070] Step 2: Filtering the environmental parameter signal to obtain a filtered environmental parameter signal;

[0071] Since the environmental parameter signal of the present invention is obtained by wireless communication, it is easily affected by external electromagnetic interference. In order to ensure the accuracy of the environmental parameters, the present invention needs to filter the environmental parameter signal.

[0072] Furthermore, step 2 includes:

[0073] Step 2.1: performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients;

[0074] Step 2.2: Calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is:

[0075]

[0076] Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale;

[0077] Step 2.3: construct a self-adjusting threshold according to the noise standard deviation;

[0078] The value of the wavelet threshold has an important influence on the final denoising effect. The wavelet threshold should be able to separate the wavelet coefficients of the noise and the original signal to the greatest extent. As the number of wavelet decomposition layers increases, the wavelet coefficient area where the noise is located should gradually decrease. Most of the wavelet coefficients decomposed at a higher level come from the original signal. Therefore, the present invention constructs a self-adjusting threshold based on the decomposition scale, and the specific process is as follows:

[0079] Specifically, the step 2.3 includes:

[0080] Using the formula:

[0081] δ i =In(e+2 1-i -1)×σ i

[0082] Construct a self-regulating threshold; where δ i Represents the self-adjustment threshold at the i-th decomposition scale.

[0083] Step 2.4: construct a self-adjusting filter function according to the self-adjusting threshold; wherein the self-adjusting filter function is:

[0084]

[0085] Construct a self-regulating filter function; where W δi represents the wavelet coefficient after filtering, W ij represents the jth wavelet coefficient under the i-th decomposition scale, k represents the adjustment factor, T represents the adaptive coefficient, and its value range is [1, 3].

[0086] Step 2.5: Filter the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal.

[0087] The present invention uses a self-adjusting filter function to perform a self-adjusting threshold transformation on the wavelet coefficients at each scale, so that the wavelet coefficients at each scale can be screened to make the final environmental parameters more accurate.

[0088] Step 3: Analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration;

[0089] Step 4: construct an outlier screening matrix based on the environmental parameters;

[0090] Further, the step 4 comprises:

[0091] Step 4.1: Construct the trust characteristic equation of each sensor according to the variance of the environmental parameter; wherein the trust characteristic equation is:

[0092]

[0093] In the formula, p i (x / x i ) represents the trust characteristic equation of the i-th sensor, x i represents the environmental parameters collected by the i-th sensor at time t, σ i represents the standard deviation of the environmental parameters collected by sensor i;

[0094] Step 4.2: Construct an outlier screening matrix based on the trust characteristic equation. In an embodiment of the present invention, the outlier screening matrix is:

[0095]

[0096] Construct an outlier screening matrix; where, represents the distance measure between sensor i and sensor j.

[0097] Step 5: Use the outlier screening matrix to remove outliers to obtain a multi-source heterogeneous data fusion set;

[0098] Specifically, the step 5 includes:

[0099] Step 5.1: removing the element values ​​in the outlier screening matrix that are greater than a preset threshold value to obtain a credibility matrix;

[0100] Step 5.2: Extract the corresponding environmental parameters in the credibility matrix to obtain a multi-source heterogeneous data fusion set.

[0101] Step 6: Fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value; it should be noted that the present invention may adopt the formula:

[0102]

[0103] The data in the multi-source heterogeneous data fusion set is integrated to obtain the construction risk assessment value; among them, represents the construction risk assessment value, W i represents the fusion weight of the i-th sensor, X i Indicates the data collected by the ith sensor in the multi-source heterogeneous data fusion set. Further, the fusion weight in the present invention can be determined according to the variance value collected by the same-source sensor. Generally, when the variance value is smaller, the fusion weight will become larger.

[0104] Step 7: When the construction risk assessment value is greater than the alarm threshold, a construction risk warning is issued.

[0105] The present invention can make the environmental parameters collected by the sensor more accurate by analyzing the filtered environmental parameter signal to obtain the environmental parameters, and then fuse all the environmental parameters based on the multi-source heterogeneous data fusion set to obtain the construction risk assessment value, which can comprehensively consider various factors to improve the accuracy of the alarm information.

[0106] See also Figure 2 The present invention also provides a dynamic risk assessment system integrating multi-source heterogeneous data, including:

[0107] An environmental parameter signal acquisition module is used to acquire environmental parameter signals collected by various sensors at the construction site;

[0108] A filtering module, used for filtering the environmental parameter signal to obtain a filtered environmental parameter signal;

[0109] An environmental parameter analysis module, used to analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration;

[0110] A screening matrix construction module, used to construct an outlier screening matrix according to the environmental parameters;

[0111] A data fusion set screening module, used for removing outliers by using the outlier screening matrix to obtain a multi-source heterogeneous data fusion set;

[0112] A fusion module, used for fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value;

[0113] The early warning module is used to issue a construction risk early warning when the construction risk assessment value is greater than the alarm threshold.

[0114] Preferably, the filtering module comprises:

[0115] A wavelet decomposition unit, used for performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients;

[0116] A noise standard deviation calculation unit is used to calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is:

[0117]

[0118] Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale;

[0119] A self-adjusting threshold value construction unit, used to construct a self-adjusting threshold value according to the noise standard deviation;

[0120] A self-adjusting filter function construction unit, used to construct a self-adjusting filter function according to the self-adjusting threshold;

[0121] A filtering unit is used to filter the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal.

[0122] The beneficial effect of a dynamic risk assessment method and system for fusing multi-source heterogeneous data provided by the present invention is that: compared with the prior art, the present invention can make the environmental parameters collected by the sensor more accurate by analyzing the filtered environmental parameter signals to obtain environmental parameters, and then fuse all environmental parameters based on the multi-source heterogeneous data fusion set to obtain a construction risk assessment value, which can comprehensively consider multiple factors to improve the accuracy of the alarm information.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dynamic risk assessment method integrating multi-source heterogeneous data, characterized in that: include: Step 1: Obtain environmental parameter signals collected by various sensors at the construction site; Step 2: Filtering the environmental parameter signal to obtain a filtered environmental parameter signal; The step 2: filtering the environmental parameter signal to obtain a filtered environmental parameter signal, comprises: Step 2.1: performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients; Step 2.2: Calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is: Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale; Step 2.3: construct a self-adjusting threshold according to the noise standard deviation; The step 2.3: constructing a self-adjusting threshold according to the noise standard deviation includes: Using the formula: δ i =In(e+2 1-i -1)×σ i Construct a self-regulating threshold; where δ i represents the self-adjustment threshold at the i-th decomposition scale; Step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold; The step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold value comprises: Using the formula: Construct a self-regulating filter function; where W δi represents the wavelet coefficient after filtering, W ij represents the jth wavelet coefficient at the i-th decomposition scale, k represents the adjustment factor, T represents the adaptive coefficient, and its value range is [1, 3]; Step 2.5: Filtering the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal; Step 3: Analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration; Step 4: construct an outlier screening matrix based on the environmental parameters; Step 5: Use the outlier screening matrix to remove outliers to obtain a multi-source heterogeneous data fusion set; Step 6: Fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value; Step 7: When the construction risk assessment value is greater than the alarm threshold, a construction risk warning is issued.

2. A dynamic risk assessment method for integrating multi-source heterogeneous data as claimed in claim 1, characterized in that: The step 4: constructing an outlier screening matrix according to the environmental parameters, comprises: Step 4.1: Construct the trust characteristic equation of each sensor according to the variance of the environmental parameter; wherein the trust characteristic equation is: In the formula, p i (x / x i ) represents the trust characteristic equation of the ith sensor, xi represents the environmental parameters collected by the ith sensor at time t, represents the standard deviation of the environmental parameters collected by sensor i; Step 4.2: Construct an outlier screening matrix based on the trust characteristic equation.

3. A dynamic risk assessment method for integrating multi-source heterogeneous data as claimed in claim 2, characterized in that: The step 4.2: constructing an outlier screening matrix according to the trust characteristic equation includes: Using the formula: Construct an outlier screening matrix; where, represents the distance measure between sensor i and sensor j.

4. A dynamic risk assessment method for integrating multi-source heterogeneous data as claimed in claim 3, characterized in that: The step 5: using the outlier screening matrix to remove outliers to obtain a multi-source heterogeneous data fusion set includes: Step 5.1: removing the element values ​​in the outlier screening matrix that are greater than a preset threshold value to obtain a credibility matrix; Step 5.2: Extract the corresponding environmental parameters in the credibility matrix to obtain a multi-source heterogeneous data fusion set.

5. A dynamic risk assessment method for integrating multi-source heterogeneous data as claimed in claim 4, characterized in that: The step 6: fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value, including: Using the formula: The data in the multi-source heterogeneous data fusion set is integrated to obtain the construction risk assessment value; among them, represents the construction risk assessment value, W i represents the fusion weight of the i-th sensor, X i Represents the data collected by the i-th sensor in the multi-source heterogeneous data fusion set.

6. A dynamic risk assessment system integrating multi-source heterogeneous data, characterized in that: include: An environmental parameter signal acquisition module is used to obtain environmental parameter signals collected by various sensors at the construction site; A filtering module, used for filtering the environmental parameter signal to obtain a filtered environmental parameter signal; The step of filtering the environmental parameter signal to obtain a filtered environmental parameter signal includes: Step 2.1: performing wavelet decomposition on the environmental parameter signal to obtain a plurality of wavelet coefficients; Step 2.2: Calculate the noise standard deviation according to the wavelet coefficients; wherein the noise standard deviation calculation formula is: Among them, σ i represents the noise standard deviation, median(X i ) represents the median value of high-frequency wavelet coefficients at the i-th decomposition scale; Step 2.3: construct a self-adjusting threshold according to the noise standard deviation; The step 2.3: constructing a self-adjusting threshold according to the noise standard deviation includes: Using the formula: δ i =In(e+2 1-i -1)×σ i Construct a self-regulating threshold; where δ i represents the self-adjustment threshold at the i-th decomposition scale; Step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold; The step 2.4: constructing a self-adjusting filter function according to the self-adjusting threshold value comprises: Using the formula: Construct a self-regulating filter function; where W δi represents the wavelet coefficient after filtering, W ij represents the jth wavelet coefficient at the i-th decomposition scale, k represents the adjustment factor, T represents the adaptive coefficient, and its value range is [1, 3]; Step 2.5: Filtering the environmental parameter signal using the self-adjusting filter function to obtain a filtered environmental parameter signal; An environmental parameter analysis module, used to analyze the filtered environmental parameter signal to obtain environmental parameters; the environmental parameters include temperature, humidity, carbon monoxide concentration, dust concentration and hydrogen sulfide concentration; A screening matrix construction module, used to construct an outlier screening matrix according to the environmental parameters; A data fusion set screening module, used for removing outliers by using the outlier screening matrix to obtain a multi-source heterogeneous data fusion set; A fusion module, used for fusing the data in the multi-source heterogeneous data fusion set to obtain a construction risk assessment value; The early warning module is used to issue a construction risk early warning when the construction risk assessment value is greater than the alarm threshold.

Images