To investigate the characteristics and sources of potentially toxic elemental contamination in soils around the coal-fired power plant, we selected the soil around the Jingyuan power plant as the object of investigation. Thirty-six and 27 soil samples collected from the urban and farmland areas were analyzed for PTEs (As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn), respectively. We employed the geoaccumulation index, single-factor pollution index, and improved Nemerow integrated pollution index to evaluate the contamination characteristics of PTEs, and combined correlation analysis with absolute principal component scores-multiple linear regression (APCS-MLR) and positive matrix factorization (PMF) receptor models were used to quantitatively analyze the pollution sources of PTEs. The results revealed that all the average concentrations of other soil metals around Jingyuan power plant exceeded their corresponding background values of Gansu Province, except for As, in agricultural soil, and Hg exhibited significant spatial non-homogeneity, and thus it was visibly affected by anthropogenic activities. The values of the single-factor pollution and geoaccumulation indices indicated that the soils of both sites were mainly contaminated with Cd, Cr, Ni, and Pb, and Hg contamination was reflected in wide areas for both sites. Additionally, the results of the improved Nemerow index demonstrated that the investigated soil was between moderately contaminated and heavily contaminated. Further, the comprehensive pollution degree of urban soil was higher than that of agricultural soil. Moreover, source apportionment revealed that PTEs in urban soil were derived from mixed sources of traffic activities and coal combustion, mixed sources of traffic activities and industrial and mining activities, and atmospheric deposition from industrial activities, with contribution rates of APCS-MLR of 35.2%, 25.1%, and 23.4%, respectively. The PMF contribution rates were 40.2%, 12.4%, and 47.7%, respectively. PTEs in agricultural soil were from one mixture source of industrial, mining, agricultural, and traffic sources, and the other mixed source was of traffic activities and coal combustion, with a APCS-MLR contribution rate of 40.3% and 35.9% and a contribution rate of PMF of 36.2% and 18.0%, respectively. In addition, another mixed source of agricultural activities and coal combustion was obtained using the PMF model, with a contribution rate of 48.5%.