Microphone Array Impact Localization for Machinery Wear Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to effectively detect and spatially localize high impact soundwaves in complex industrial machinery, leading to inefficiencies and downtime due to wear phenomena like liner wear in ball mills, as they are not designed to handle transient, non-stationary, and wideband sound signals, and conventional methods increase network complexity.
Innovation Solution
A method and system using a microphone array to record sound signals, apply filtering, perform eigenvalue decomposition, and determine a spatial power matrix to localize impact hotspots by projecting a spatial localization map onto the machinery surface, identifying peaks and accumulating impact hotspots over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a two-stage approach with separate networks is used for sound event detection and direction-of-arrival estimation, then the system avoids multi-objective optimization problems, but the device complexity and network size increase significantly
Solution Approach 1:
The patent combines sound event detection and direction-of-arrival estimation into a single unified neural network that processes audio signals simultaneously for both tasks. This integration reduces the system from multiple separate networks to one cohesive architecture, decreasing computational overhead and device complexity while maintaining detection reliability through joint optimization of both functions.
2Device complexity
If conventional sound detection methods are used, then the system structure is simple, but the system fails to detect transient, non-stationary, and wideband impact soundwaves in complex machinery
Solution Approach 1:
The patent employs a unified neural network with dynamically adjustable parameters designed specifically to handle transient, non-stationary, and wideband characteristics of impact soundwaves. The network adapts its processing parameters to capture high-frequency impact signals in complex machinery environments, achieving precise impact detection while maintaining a relatively simple overall system structure through parameter optimization rather than structural complexity.
3Ease of operation
If manual inspection and abrupt stoppage are used for machinery maintenance, then the system is simple to operate, but huge downtime losses occur
Solution Approach 1:
The patent replaces manual inspection methods with an automated acoustic monitoring system using a unified neural network that continuously analyzes sound signals from machinery. This substitution enables real-time detection of impact events and predictive maintenance scheduling, eliminating the need for abrupt stoppages and manual inspections, thereby reducing downtime losses while maintaining ease of operation through automated monitoring.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Accurately localizes impact sound sources with less than 10% root mean square error (RMSE), enabling predictive maintenance by identifying wear zones without contact-based sensing, suitable for planar and curved surfaces.
Implementation Method 1
placing a microphone array at different locations within the machinery to record a sound signal
Implementation Method 2
The the time windowed sound signal is filtered by applying filtering (308) to obtain a filtered sound signal based on (i) a sliding narrow band pass filter of central frequency f, (ii) a frequency range between a spectral response of impact type, and (iii) a band distance
Implementation Method 3
The method performs an eigenvalue decomposition over a covariance matrix of the filtered sound signal s(t)FILT to determine a noise subspace
Implementation Method 4
a steering vector is obtained from the geometrical structure parameters of the machinery and the noise subspace to evaluate a spatial power matrix for each time window of the sound signal. Then, a spatial localization map is obtained on a microphone array plane for each time window using the spatial power matrix
Data Source
AI summary
This disclosure relates generally to method and system for unobtrusive spatial localization and detection of impact induced soundwaves. Runtime localization of surface wear zones in industrial machines has been a daunting problem in the domain of predictive maintenance. The method disclosed provides spatial localization of impacts soundwaves generated to locate the wearzones. The method processes the sound signal from the microphone array occurred surface of target of interest of the machinery. The sound signal is processed to identify one or more impact hotspots on the target surface of the machinery by searching one or more peaks. Finally, exact location of occurred impact sound from the one or more impact hotspots is localized and is projected on the machinery.